[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"similar-unslothai--notebooks":3,"tool-unslothai--notebooks":61},[4,18,26,36,44,53],{"id":5,"name":6,"github_repo":7,"description_zh":8,"stars":9,"difficulty_score":10,"last_commit_at":11,"category_tags":12,"status":17},4358,"openclaw","openclaw\u002Fopenclaw","OpenClaw 是一款专为个人打造的本地化 AI 助手，旨在让你在自己的设备上拥有完全可控的智能伙伴。它打破了传统 AI 助手局限于特定网页或应用的束缚，能够直接接入你日常使用的各类通讯渠道，包括微信、WhatsApp、Telegram、Discord、iMessage 等数十种平台。无论你在哪个聊天软件中发送消息，OpenClaw 都能即时响应，甚至支持在 macOS、iOS 和 Android 设备上进行语音交互，并提供实时的画布渲染功能供你操控。\n\n这款工具主要解决了用户对数据隐私、响应速度以及“始终在线”体验的需求。通过将 AI 部署在本地，用户无需依赖云端服务即可享受快速、私密的智能辅助，真正实现了“你的数据，你做主”。其独特的技术亮点在于强大的网关架构，将控制平面与核心助手分离，确保跨平台通信的流畅性与扩展性。\n\nOpenClaw 非常适合希望构建个性化工作流的技术爱好者、开发者，以及注重隐私保护且不愿被单一生态绑定的普通用户。只要具备基础的终端操作能力（支持 macOS、Linux 及 Windows WSL2），即可通过简单的命令行引导完成部署。如果你渴望拥有一个懂你",349277,3,"2026-04-06T06:32:30",[13,14,15,16],"Agent","开发框架","图像","数据工具","ready",{"id":19,"name":20,"github_repo":21,"description_zh":22,"stars":23,"difficulty_score":10,"last_commit_at":24,"category_tags":25,"status":17},3808,"stable-diffusion-webui","AUTOMATIC1111\u002Fstable-diffusion-webui","stable-diffusion-webui 是一个基于 Gradio 构建的网页版操作界面，旨在让用户能够轻松地在本地运行和使用强大的 Stable Diffusion 图像生成模型。它解决了原始模型依赖命令行、操作门槛高且功能分散的痛点，将复杂的 AI 绘图流程整合进一个直观易用的图形化平台。\n\n无论是希望快速上手的普通创作者、需要精细控制画面细节的设计师，还是想要深入探索模型潜力的开发者与研究人员，都能从中获益。其核心亮点在于极高的功能丰富度：不仅支持文生图、图生图、局部重绘（Inpainting）和外绘（Outpainting）等基础模式，还独创了注意力机制调整、提示词矩阵、负向提示词以及“高清修复”等高级功能。此外，它内置了 GFPGAN 和 CodeFormer 等人脸修复工具，支持多种神经网络放大算法，并允许用户通过插件系统无限扩展能力。即使是显存有限的设备，stable-diffusion-webui 也提供了相应的优化选项，让高质量的 AI 艺术创作变得触手可及。",162132,"2026-04-05T11:01:52",[14,15,13],{"id":27,"name":28,"github_repo":29,"description_zh":30,"stars":31,"difficulty_score":32,"last_commit_at":33,"category_tags":34,"status":17},1381,"everything-claude-code","affaan-m\u002Feverything-claude-code","everything-claude-code 是一套专为 AI 编程助手（如 Claude Code、Codex、Cursor 等）打造的高性能优化系统。它不仅仅是一组配置文件，而是一个经过长期实战打磨的完整框架，旨在解决 AI 代理在实际开发中面临的效率低下、记忆丢失、安全隐患及缺乏持续学习能力等核心痛点。\n\n通过引入技能模块化、直觉增强、记忆持久化机制以及内置的安全扫描功能，everything-claude-code 能显著提升 AI 在复杂任务中的表现，帮助开发者构建更稳定、更智能的生产级 AI 代理。其独特的“研究优先”开发理念和针对 Token 消耗的优化策略，使得模型响应更快、成本更低，同时有效防御潜在的攻击向量。\n\n这套工具特别适合软件开发者、AI 研究人员以及希望深度定制 AI 工作流的技术团队使用。无论您是在构建大型代码库，还是需要 AI 协助进行安全审计与自动化测试，everything-claude-code 都能提供强大的底层支持。作为一个曾荣获 Anthropic 黑客大奖的开源项目，它融合了多语言支持与丰富的实战钩子（hooks），让 AI 真正成长为懂上",159267,2,"2026-04-17T11:29:14",[14,13,35],"语言模型",{"id":37,"name":38,"github_repo":39,"description_zh":40,"stars":41,"difficulty_score":32,"last_commit_at":42,"category_tags":43,"status":17},2271,"ComfyUI","Comfy-Org\u002FComfyUI","ComfyUI 是一款功能强大且高度模块化的视觉 AI 引擎，专为设计和执行复杂的 Stable Diffusion 图像生成流程而打造。它摒弃了传统的代码编写模式，采用直观的节点式流程图界面，让用户通过连接不同的功能模块即可构建个性化的生成管线。\n\n这一设计巧妙解决了高级 AI 绘图工作流配置复杂、灵活性不足的痛点。用户无需具备编程背景，也能自由组合模型、调整参数并实时预览效果，轻松实现从基础文生图到多步骤高清修复等各类复杂任务。ComfyUI 拥有极佳的兼容性，不仅支持 Windows、macOS 和 Linux 全平台，还广泛适配 NVIDIA、AMD、Intel 及苹果 Silicon 等多种硬件架构，并率先支持 SDXL、Flux、SD3 等前沿模型。\n\n无论是希望深入探索算法潜力的研究人员和开发者，还是追求极致创作自由度的设计师与资深 AI 绘画爱好者，ComfyUI 都能提供强大的支持。其独特的模块化架构允许社区不断扩展新功能，使其成为当前最灵活、生态最丰富的开源扩散模型工具之一，帮助用户将创意高效转化为现实。",108322,"2026-04-10T11:39:34",[14,15,13],{"id":45,"name":46,"github_repo":47,"description_zh":48,"stars":49,"difficulty_score":32,"last_commit_at":50,"category_tags":51,"status":17},6121,"gemini-cli","google-gemini\u002Fgemini-cli","gemini-cli 是一款由谷歌推出的开源 AI 命令行工具，它将强大的 Gemini 大模型能力直接集成到用户的终端环境中。对于习惯在命令行工作的开发者而言，它提供了一条从输入提示词到获取模型响应的最短路径，无需切换窗口即可享受智能辅助。\n\n这款工具主要解决了开发过程中频繁上下文切换的痛点，让用户能在熟悉的终端界面内直接完成代码理解、生成、调试以及自动化运维任务。无论是查询大型代码库、根据草图生成应用，还是执行复杂的 Git 操作，gemini-cli 都能通过自然语言指令高效处理。\n\n它特别适合广大软件工程师、DevOps 人员及技术研究人员使用。其核心亮点包括支持高达 100 万 token 的超长上下文窗口，具备出色的逻辑推理能力；内置 Google 搜索、文件操作及 Shell 命令执行等实用工具；更独特的是，它支持 MCP（模型上下文协议），允许用户灵活扩展自定义集成，连接如图像生成等外部能力。此外，个人谷歌账号即可享受免费的额度支持，且项目基于 Apache 2.0 协议完全开源，是提升终端工作效率的理想助手。",100752,"2026-04-10T01:20:03",[52,13,15,14],"插件",{"id":54,"name":55,"github_repo":56,"description_zh":57,"stars":58,"difficulty_score":32,"last_commit_at":59,"category_tags":60,"status":17},4721,"markitdown","microsoft\u002Fmarkitdown","MarkItDown 是一款由微软 AutoGen 团队打造的轻量级 Python 工具，专为将各类文件高效转换为 Markdown 格式而设计。它支持 PDF、Word、Excel、PPT、图片（含 OCR）、音频（含语音转录）、HTML 乃至 YouTube 链接等多种格式的解析，能够精准提取文档中的标题、列表、表格和链接等关键结构信息。\n\n在人工智能应用日益普及的今天，大语言模型（LLM）虽擅长处理文本，却难以直接读取复杂的二进制办公文档。MarkItDown 恰好解决了这一痛点，它将非结构化或半结构化的文件转化为模型“原生理解”且 Token 效率极高的 Markdown 格式，成为连接本地文件与 AI 分析 pipeline 的理想桥梁。此外，它还提供了 MCP（模型上下文协议）服务器，可无缝集成到 Claude Desktop 等 LLM 应用中。\n\n这款工具特别适合开发者、数据科学家及 AI 研究人员使用，尤其是那些需要构建文档检索增强生成（RAG）系统、进行批量文本分析或希望让 AI 助手直接“阅读”本地文件的用户。虽然生成的内容也具备一定可读性，但其核心优势在于为机器",93400,"2026-04-06T19:52:38",[52,14],{"id":62,"github_repo":63,"name":64,"description_en":65,"description_zh":66,"ai_summary_zh":66,"readme_en":67,"readme_zh":68,"quickstart_zh":69,"use_case_zh":70,"hero_image_url":71,"owner_login":72,"owner_name":73,"owner_avatar_url":74,"owner_bio":75,"owner_company":76,"owner_location":76,"owner_email":76,"owner_twitter":72,"owner_website":77,"owner_url":78,"languages":79,"stars":92,"forks":93,"last_commit_at":94,"license":95,"difficulty_score":32,"env_os":96,"env_gpu":97,"env_ram":98,"env_deps":99,"category_tags":111,"github_topics":114,"view_count":32,"oss_zip_url":76,"oss_zip_packed_at":76,"status":17,"created_at":115,"updated_at":116,"faqs":117,"releases":153},8539,"unslothai\u002Fnotebooks","notebooks","250+ Fine-tuning & RL Notebooks for text, vision, audio, embedding, TTS models.","notebooks 是 Unsloth 团队精心打造的开源项目，汇集了超过 250 个针对文本、视觉、音频、嵌入模型及语音合成（TTS）的微调与强化学习（RL）实战笔记。它主要解决了 AI 开发者在模型定制过程中面临的环境配置复杂、代码复用难以及从数据准备到推理全流程打通门槛高等痛点。\n\n无论是希望快速上手的初学者，还是需要高效实验的资深研究人员与工程师，都能从中受益。用户无需从零编写底层代码，即可直接基于 Google Colab 或本地环境，对 Gemma、Qwen 等主流大模型进行可视化微调、对话训练甚至复杂的 GRPO 强化学习。\n\n其核心亮点在于“开箱即用”的完整性：每个笔记都涵盖了数据预处理、模型训练到最终推理的全部环节，并针对 Unsloth 的高性能优化特性进行了适配。通过提供清晰的模型分类索引和一键运行链接，notebooks 极大地缩短了算法验证周期，让创作者能更专注于策略调整与应用创新，而非繁琐的工程实现细节。","\u003Cdiv align=\"center\">\n\n \u003Ca href=\"https:\u002F\u002Funsloth.ai\">\u003Cpicture>\n \u003Csource media=\"(prefers-color-scheme: dark)\" srcset=\"https:\u002F\u002Fraw.githubusercontent.com\u002Funslothai\u002Funsloth\u002Fmain\u002Fimages\u002Funsloth%20logo%20white%20text.png\">\n \u003Csource media=\"(prefers-color-scheme: light)\" srcset=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Funslothai_notebooks_readme_f98b20d7f469.png\">\n \u003Cimg alt=\"unsloth logo\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Funslothai_notebooks_readme_f98b20d7f469.png\" height=\"110\" style=\"max-width: 100%;\">\n \u003C\u002Fpicture>\u003C\u002Fa>\n \n\u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002Fgpt-oss-(20B)-Fine-tuning.ipynb\">\u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Funslothai_notebooks_readme_6e935a75881d.png\" height=\"48\">\u003C\u002Fa>\n\u003Ca href=\"https:\u002F\u002Fdiscord.gg\u002Funsloth\">\u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Funslothai_notebooks_readme_a85808c43728.png\" height=\"48\">\u003C\u002Fa>\n\u003Ca href=\"https:\u002F\u002Funsloth.ai\u002Fdocs\u002Fget-started\u002Funsloth-notebooks\">\u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Funslothai_notebooks_readme_cb417d7bd7e4.png\" height=\"48\">\u003C\u002Fa>\n\n\u003C\u002Fdiv>\n\n## 📒 Fine-tuning Notebooks\nBelow are Colab notebooks, organized by model. You can also view all [notebooks in our docs](https:\u002F\u002Funsloth.ai\u002Fdocs\u002Fget-started\u002Funsloth-notebooks).\u003Cbr>The notebooks run locally and feature data prep, training and inference. Read our [fine-tuning guide](https:\u002F\u002Funsloth.ai\u002Fdocs\u002Fget-started\u002Ffine-tuning-llms-guide).\n\n### Main Notebooks\n| Model | Type | Notebook Link |\n|-----------------------------|----------------|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|\n| **Unsloth Studio** | Chat UI | [![Open in Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FUnsloth_Studio.ipynb) |\n| **Gemma 4 (E2B)** | Vision | [![Open in Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma4_%28E2B%29-Vision.ipynb) |\n| **Qwen3.5 (4B)** | Vision | [![Open in Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_5_%284B%29_Vision.ipynb) |\n| **Qwen3.5 (2B)** | Vision | [![Open in Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_5_%282B%29_Vision.ipynb) |\n| **gpt-oss (20B)** | Fine-tuning | [![Open in Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002Fgpt-oss-%2820B%29-Fine-tuning.ipynb) |\n| **gpt-oss (20B)** | GRPO | [![Open in Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002Fgpt-oss-%2820B%29-GRPO.ipynb) |\n| **Qwen3 (14B)** | Conversational | [![Open in Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_%2814B%29-Reasoning-Conversational.ipynb) |\n| **Qwen3-VL (8B)** | Vision | [![Open in Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_VL_%288B%29-Vision.ipynb) |\n| **Qwen3-Embedding (0.6B)** | Embeddings | [![Open in Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_Embedding_(0_6B).ipynb) |\n| **Qwen3: Advanced GRPO** | GRPO | [![Open in Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_%284B%29-GRPO.ipynb) |\n| **Gemma 3 (4B)** | Vision | [![Open in Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma3_%284B%29-Vision.ipynb) |\n| **Gemma 3N (4B)** | Audio | [![Open in Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma3N_%284B%29-Audio.ipynb) |\n| **embeddinggemma (300M)** | Embeddings | [![Open in Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FEmbeddingGemma_%28300M%29.ipynb) |\n| **Mistral Ministral 3 (3B)**| Vision | [![Open in Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FMinistral_3_VL_%283B%29_Vision.ipynb) |\n| **Mistral v0.3 (7B)** | Vision | [![Open in Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FMistral_v0.3_(7B)-Alpaca.ipynb) |\n| **Llama 3.1 (8B) Alpaca** | Alpaca | [![Open in Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLlama3.1_%288B%29-Alpaca.ipynb) |\n| **Llama 3.2 (1B + 3B)**| Conversational | [![Open in Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLlama3.2_%281B_and_3B%29-Conversational.ipynb) |\n| **Phi-4 (14B)** | Conversational | [![Open in Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FPhi_4-Conversational.ipynb) |\n| **Orpheus-TTS (3B)** | TTS | [![Open in Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FOrpheus_%283B%29-TTS.ipynb) |\n\n\u003C!-- 🛑 🚨 DO NOT EDIT MANUALLY THIS SECTION UNTIL `end of notebook links`!! 🛑 🚨 -->\n\u003C!-- 🛑 🚨 THIS SECTION IS GENERATED BY `update_all_notebooks.py` AUTOMATICALLY 🛑 🚨 -->\n\n\u003C!-- START OF EDITING -->\n### GRPO & Reinforcement Learning Notebooks\n| Model | Type | Notebook Link |\n| --- | --- | --- |\n| **Llama3.1** **(8B)** | GSM8K Math + vLLM | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLlama3.1_(8B)-GRPO.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **NeMo Gym Sudoku** | Sudoku | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FNeMo-Gym-Sudoku.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **NeMo Gym Multi Environment** | Multi Environment | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FNeMo-Gym-Multi-Environment.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **gpt oss BF16** **(20B)** | 2048 Game | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002Fgpt_oss_(20B)_Reinforcement_Learning_2048_Game_BF16.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **gpt oss** **(20B)** | Minesweeper Game | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002Fgpt_oss_(20B)_Reinforcement_Learning_GRPO_Minesweeper_Game_BF16.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **gpt oss** **(20B)** | Auto Kernel Creation | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002Fgpt_oss_(20B)_GRPO_BF16.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Qwen3** **(8B)** | DAPO Math + vLLM | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_8B_FP8_GRPO.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Llama3** **(8B)** | ORPO | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLlama3_(8B)-ORPO.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Openenv wordle** | Wordle + vLLM | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FOpenenv_wordle_grpo.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Qwen2.5 VL** **(7B)** | Vision Math + vLLM | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen2_5_7B_VL_GRPO.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Zephyr** **(7B)** | DPO | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FZephyr_(7B)-DPO.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **(OpenEnv) gpt oss BF16** **(20B)** | 2048 Game | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FOpenEnv_gpt_oss_(20B)_Reinforcement_Learning_2048_Game_BF16.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **gpt oss** **(20B)** | Auto Kernel Creation | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002Fgpt-oss-(20B)-GRPO.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **gpt oss** **(20B)** | 2048 Game | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002Fgpt_oss_(20B)_Reinforcement_Learning_2048_Game.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **(OpenEnv) gpt oss** **(20B)** | 2048 Game | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FOpenEnv_gpt_oss_(20B)_Reinforcement_Learning_2048_Game.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **(DGX Spark) gpt oss** **(20B)** | 2048 Game | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002Fgpt_oss_(20B)_Reinforcement_Learning_2048_Game_DGX_Spark.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **(A100) gpt oss** **(20B)** | Auto Kernel Creation | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002Fgpt-oss-(20B)_A100-GRPO.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Qwen2.5** **(3B)** | GSM8K Math + vLLM | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen2.5_(3B)-GRPO.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Llama3.2** **(1B)** | DAPO Math + vLLM | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLlama_FP8_GRPO.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Qwen3 VL** **(8B)** | Vision Math | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_VL_(8B)-Vision-GRPO.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Mistral v0.3** **(7B)** | GSM8K Math + vLLM | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FMistral_v0.3_(7B)-GRPO.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Qwen3 5** **(4B)** | Vision Math | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_5_(4B)_Vision_GRPO.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Gemma3** **(4B)** | Vision Math | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma3_(4B)-Vision-GRPO.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Phi 4** **(14B)** | GSM8K Math + vLLM | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FPhi_4_(14B)-GRPO.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Gemma3** **(1B)** | GSM8K Math | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma3_(1B)-GRPO.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **DeepSeek R1 0528 Qwen3** **(8B)** | DAPO Math + vLLM | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FDeepSeek_R1_0528_Qwen3_(8B)_GRPO.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **LFM2.5** **(1.2B)** | DAPO Math | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLFM2.5_(1.2B)-GRPO.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Qwen3** **(4B)** | DAPO Math + vLLM | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_(4B)-GRPO.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Ministral3** **(3B)** | Sudoku | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FMinistral_3_(3B)_Reinforcement_Learning_Sudoku_Game.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n\n### Tool Calling Notebooks\n| Model | Type | Notebook Link |\n| --- | --- | --- |\n| **Qwen2.5 Coder** **(1.5B)** | Tool Calling | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen2.5_Coder_(1.5B)-Tool_Calling.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **FunctionGemma** **(270M)** | Tool Calling | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FFunctionGemma_(270M)-Multi-Turn-Tool-Calling.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **FunctionGemma** **(270M)** | Mobile Actions | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FFunctionGemma_(270M)-Mobile-Actions.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **FunctionGemma** **(270M)** | Inference | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FFunctionGemma_(270M)-LMStudio.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **FunctionGemma** **(270M)** | Conversational | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FFunctionGemma_(270M).ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n\n### Text-to-Speech (TTS) Notebooks\n| Model | Type | Notebook Link |\n| --- | --- | --- |\n| **Spark TTS** **(0.5B)** | TTS | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FSpark_TTS_(0_5B).ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Orpheus** **(3B)** | TTS | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FOrpheus_(3B)-TTS.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Llasa TTS** **(3B)** | TTS | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLlasa_TTS_(3B).ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Llasa TTS** **(1B)** | TTS | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLlasa_TTS_(1B).ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Sesame CSM** **(1B)** | TTS | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FSesame_CSM_(1B)-TTS.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Oute TTS** **(1B)** | TTS | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FOute_TTS_(1B).ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n\n### Vision (Multimodal) Notebooks\n| Model | Type | Notebook Link |\n| --- | --- | --- |\n| **Qwen2.5 VL** **(7B)** | Vision Math + vLLM (GRPO RL) | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen2_5_7B_VL_GRPO.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Gemma4** **(E4B)** | Vision | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma4_(E4B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Gemma4** **(E2B)** | Vision | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma4_(E2B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Gemma4** **(26B A4B)** | Vision | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma4_(26B_A4B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Gemma4** **(31B)** | Vision | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma4_(31B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Qwen2.5 VL** **(7B)** | Vision | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen2.5_VL_(7B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **ERNIE 4 5 VL 28B A3B PT** | Vision | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FERNIE_4_5_VL_28B_A3B_PT_Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Qwen3 VL** **(8B)** | Vision | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_VL_(8B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Qwen3 VL** **(8B)** | Vision Math (GRPO RL) | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_VL_(8B)-Vision-GRPO.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Qwen3 5** **(4B)** | Vision | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_5_(4B)_Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Qwen3 5** **(4B)** | Vision Math (GRPO RL) | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_5_(4B)_Vision_GRPO.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Gemma3** **(4B)** | Vision Math (GRPO RL) | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma3_(4B)-Vision-GRPO.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Qwen3 5** **(0 8B)** | Vision | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_5_(0_8B)_Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Qwen2 VL** **(7B)** | Vision | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen2_VL_(7B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Llama3.2** **(11B)** | Vision | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLlama3.2_(11B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Qwen3 5** **(2B)** | Vision | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_5_(2B)_Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Pixtral** **(12B)** | Vision | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FPixtral_(12B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **LFM2.5 VL** **(1.6B)** | Vision | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLFM2.5_VL_(1.6B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Ministral3 VL** **(3B)** | Vision | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FMinistral_3_VL_(3B)_Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Gemma3** **(4B)** | Vision | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma3_(4B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Gemma3N** **(4B)** | Vision | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma3N_(4B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n\n### Embedding Notebooks\n| Model | Type | Notebook Link |\n| --- | --- | --- |\n| **ModernBert** | | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FModernBert.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Qwen3 Embedding** **(4B)** | | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_Embedding_(4B).ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Qwen3 Embedding** **(0 6B)** | | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_Embedding_(0_6B).ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **EmbeddingGemma** **(300M)** | | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FEmbeddingGemma_(300M).ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **BGE M3** | | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FBGE_M3.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **ModernBERT** **(Large)** | Classification | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002Fbert_classification.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **All MiniLM L6 v2** | | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FAll_MiniLM_L6_v2.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n\n### Speech-to-Text (STT) Notebooks\n| Model | Type | Notebook Link |\n| --- | --- | --- |\n| **Whisper** **(Large)** | Fine Tuning | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FWhisper.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n\n### OCR Notebooks\n| Model | Type | Notebook Link |\n| --- | --- | --- |\n| **Deepseek OCR** **(3B)** | Fine Tuning | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FDeepseek_OCR_(3B).ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Deepseek OCR** **(3B)** | Evaluation | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FDeepseek_OCR_(3B)-Evaluation.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Deepseek OCR** **(3B)** | Eval | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FDeepseek_OCR_(3B)-Eval.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Deepseek OCR 2** **(3B)** | | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FDeepseek_OCR_2_(3B).ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Paddle OCR** **(1B)** | Vision | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FPaddle_OCR_(1B)_Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n\n### BERT Notebooks\n| Model | Type | Notebook Link |\n| --- | --- | --- |\n| **ModernBert** | | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FModernBert.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **ModernBERT** **(Large)** | Classification | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002Fbert_classification.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n\n### Deepseek Notebooks\n| Model | Type | Notebook Link |\n| --- | --- | --- |\n| **Deepseek OCR** **(3B)** | Fine Tuning | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FDeepseek_OCR_(3B).ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Deepseek OCR** **(3B)** | Evaluation | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FDeepseek_OCR_(3B)-Evaluation.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Deepseek OCR** **(3B)** | Eval | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FDeepseek_OCR_(3B)-Eval.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Deepseek OCR 2** **(3B)** | | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FDeepseek_OCR_2_(3B).ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n\n### ERNIE Notebooks\n| Model | Type | Notebook Link |\n| --- | --- | --- |\n| **ERNIE 4 5 VL 28B A3B PT** | Vision | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FERNIE_4_5_VL_28B_A3B_PT_Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **ERNIE 4 5 21B A3B PT** | Conversational | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FERNIE_4_5_21B_A3B_PT-Conversational.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n\n### GLM Notebooks\n| Model | Type | Notebook Link |\n| --- | --- | --- |\n| **(A100) GLM Flash(80GB)** | | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGLM_Flash_A100(80GB).ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n\n### GPT-OSS Notebooks\n| Model | Type | Notebook Link |\n| --- | --- | --- |\n| **gpt oss MXFP4** **(20B)** | Inference | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGPT_OSS_MXFP4_(20B)-Inference.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **gpt oss** **(20B)** | Fine Tuning | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002Fgpt_oss_(20B)_500K_Context_Fine_tuning.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **gpt oss** **(20B)** | Fine Tuning | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002Fgpt-oss-(20B)-Fine-tuning.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **gpt oss BNB** **(20B)** | Inference | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGPT_OSS_BNB_(20B)-Inference.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **(A100) gpt oss** **(120B)** | Fine Tuning | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002Fgpt-oss-(120B)_A100-Fine-tuning.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n\n### Gemma 4 Notebooks\n| Model | Type | Notebook Link |\n| --- | --- | --- |\n| **Gemma4** **(E4B)** | Vision | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma4_(E4B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Gemma4** **(E4B)** | Conversational | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma4_(E4B)-Text.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Gemma4** **(E4B)** | Audio | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma4_(E4B)-Audio.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Gemma4** **(E2B)** | Vision | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma4_(E2B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Gemma4** **(E2B)** | Audio | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma4_(E2B)-Audio.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Gemma4** **(E2B)** | Conversational | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma4_(E2B)-Text.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Gemma4** **(26B A4B)** | Conversational | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma4_(26B_A4B)-Text.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Gemma4** **(26B A4B)** | Vision | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma4_(26B_A4B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Gemma4** **(31B)** | Vision | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma4_(31B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Gemma4** **(31B)** | Conversational | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma4_(31B)-Text.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n\n### Gemma Notebooks\n| Model | Type | Notebook Link |\n| --- | --- | --- |\n| **Gemma3** **(4B)** | Conversational | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma3_(4B).ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **(A100) Gemma3** **(27B)** | Conversational | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma3_(27B)_A100-Conversational.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Gemma3** **(270M)** | Conversational | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma3_(270M).ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Gemma3** **(270M)** | | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma3_(270M)_Phone_Deployment.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Gemma3N** **(4B)** | Multimodal | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma3N_(4B)-Conversational.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Gemma3N** **(4B)** | Audio | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma3N_(4B)-Audio.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Gemma3N** **(2B)** | Inference | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma3N_(2B)-Inference.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Gemma3** **(4B)** | Vision | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma3_(4B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Gemma3N** **(4B)** | Vision | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma3N_(4B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **EmbeddingGemma** **(300M)** | | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FEmbeddingGemma_(300M).ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Gemma2** **(9B)** | Alpaca | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma2_(9B)-Alpaca.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Gemma2** **(2B)** | Alpaca | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma2_(2B)-Alpaca.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **CodeGemma** **(7B)** | Conversational | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FCodeGemma_(7B)-Conversational.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n\n### Granite Notebooks\n| Model | Type | Notebook Link |\n| --- | --- | --- |\n| **Granite4.0** **(3B)** | Conversational | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGranite4.0.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Granite4.0** **(350M)** | Conversational | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGranite4.0_350M.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n\n### Hybrid Attention Notebooks\n| Model | Type | Notebook Link |\n| --- | --- | --- |\n| **LFM2.5** **(1.2B)** | Conversational | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLFM2.5_(1.2B)-Conversational.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **LFM2.5 VL** **(1.6B)** | Vision | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLFM2.5_VL_(1.6B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Liquid LFM2** **(1.2B)** | Conversational | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLiquid_LFM2_(1.2B)-Conversational.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Liquid LFM2** | Conversational | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLiquid_LFM2-Conversational.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **LFM2.5** **(1.2B)** | | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLFM2.5_(1.2B)-Translation.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Falcon H1** **(0.5B)** | Alpaca | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FFalcon_H1_(0.5B)-Alpaca.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Falcon H1** | Alpaca | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FFalcon_H1-Alpaca.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n\n### Llama Notebooks\n| Model | Type | Notebook Link |\n| --- | --- | --- |\n| **Llama3.1** **(8B)** | Inference | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLlama3.1_(8B)-Inference.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Llama3** **(8B)** | Ollama | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLlama3_(8B)-Ollama.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Llama3** **(8B)** | Alpaca | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLlama3_(8B)-Alpaca.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Llama3.2** **(1B and 3B)** | Conversational | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLlama3.2_(1B_and_3B)-Conversational.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Llama3.2** **(1B)** | RAFT | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLlama3.2_(1B)-RAFT.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Llama3.1** **(8B)** | Alpaca | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLlama3.1_(8B)-Alpaca.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **(A100) Llama3.3** **(70B)** | Conversational | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLlama3.3_(70B)_A100-Conversational.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Llama3.2** **(11B)** | Vision | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLlama3.2_(11B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Llama3** **(8B)** | Conversational | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLlama3_(8B)-Conversational.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **TinyLlama** **(1.1B)** | Alpaca | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FTinyLlama_(1.1B)-Alpaca.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n\n### Mistral Notebooks\n| Model | Type | Notebook Link |\n| --- | --- | --- |\n| **Mistral v0.3** **(7B)** | Conversational | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FMistral_v0.3_(7B)-Conversational.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Magistral** **(24B)** | Reasoning Conversational | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FMagistral_(24B)-Reasoning-Conversational.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Pixtral** **(12B)** | Vision | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FPixtral_(12B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Mistral Small** **(22B)** | Alpaca | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FMistral_Small_(22B)-Alpaca.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Ministral3 VL** **(3B)** | Vision | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FMinistral_3_VL_(3B)_Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Mistral v0.3** **(7B)** | Alpaca | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FMistral_v0.3_(7B)-Alpaca.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Mistral Nemo** **(12B)** | Alpaca | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FMistral_Nemo_(12B)-Alpaca.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n\n### Nemotron Notebooks\n| Model | Type | Notebook Link |\n| --- | --- | --- |\n| **(A100) Nemotron Nano 3 30B A3B** | | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FNemotron-Nano-3-30B-A3B_A100.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **(A100) Nemotron 3 Nano 30B A3B** | | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FNemotron-3-Nano-30B-A3B_A100.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n\n### Paddle Notebooks\n| Model | Type | Notebook Link |\n| --- | --- | --- |\n| **Paddle OCR** **(1B)** | Vision | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FPaddle_OCR_(1B)_Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n\n### Phi Notebooks\n| Model | Type | Notebook Link |\n| --- | --- | --- |\n| **Phi 4** | Conversational | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FPhi_4-Conversational.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Phi 3.5 Mini** | Conversational | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FPhi_3.5_Mini-Conversational.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Phi 3 Medium** | Conversational | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FPhi_3_Medium-Conversational.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n\n### Qwen Notebooks\n| Model | Type | Notebook Link |\n| --- | --- | --- |\n| **Qwen3** **(0.6B)** | Reasoning Conversational | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_(0.6B)-Reasoning-Conversational-ExecuTorch.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Qwen3** **(0 6B)** | | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_(0_6B)-Phone_Deployment.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Qwen3** **(4B)** | QAT | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_(4B)_Instruct-QAT.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Qwen3** **(4B)** | Conversational | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_(4B)-Instruct.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Qwen2.5 VL** **(7B)** | Vision | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen2.5_VL_(7B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Qwen3 VL** **(8B)** | Vision | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_VL_(8B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Qwen3 5 MoE** | | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_5_MoE.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **(A100) Qwen 3 5 27B(80GB)** | | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen_3_5_27B_A100(80GB).ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **(A100) Qwen3** **(32B)** | Reasoning Conversational | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_(32B)_A100-Reasoning-Conversational.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Qwen3 5** **(4B)** | Vision | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_5_(4B)_Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Qwen3** **(14B)** | Reasoning Conversational | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_(14B)-Reasoning-Conversational.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Qwen3** **(14B)** | | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_(14B).ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Qwen3 5** **(0 8B)** | Vision | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_5_(0_8B)_Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Qwen2 VL** **(7B)** | Vision | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen2_VL_(7B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Qwen3** **(4B)** | Thinking | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_(4B)-Thinking.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Qwen3 MoE** | | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_MoE.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Qwen3 5** **(2B)** | Vision | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_5_(2B)_Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Qwen2.5** **(7B)** | Alpaca | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen2.5_(7B)-Alpaca.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Qwen2.5 Coder** **(14B)** | Conversational | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen2.5_Coder_(14B)-Conversational.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Qwen3 Embedding** **(4B)** | | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_Embedding_(4B).ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Qwen3 Embedding** **(0 6B)** | | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_Embedding_(0_6B).ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Qwen3** **(14B)** | Alpaca | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_(14B)-Alpaca.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Qwen2** **(7B)** | Alpaca | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen2_(7B)-Alpaca.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **TinyQwen3 MoE** | | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FTinyQwen3_MoE.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n\n### Text Completion \u002F Continued Pretraining Notebooks\n| Model | Type | Notebook Link |\n| --- | --- | --- |\n| **LFM2.5** **(1.2B)** | Text Completion | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLFM2.5_(1.2B)-Text_Completion.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Mistral v0.3** **(7B)** | CPT | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FMistral_v0.3_(7B)-CPT.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Mistral** **(7B)** | Text Completion | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FMistral_(7B)-Text_Completion.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n\n### Specific use-case Notebooks\n| Usecase | Model | Notebook Link |\n| --- | --- | --- |\n| Text Classification | Llama 3.1 (8B) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ftimothelaborie\u002Ftext_classification_scripts\u002Fblob\u002Fmain\u002Funsloth_classification.ipynb) |\n| Tool Calling | Qwen2.5-Coder (1.5B) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen2.5_Coder_(1.5B)-Tool_Calling.ipynb) |\n| Multiple Datasets | | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1njCCbE1YVal9xC83hjdo2hiGItpY_D6t?usp=sharing) |\n| KTO | Qwen2.5-Instruct (1.5B) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1MRgGtLWuZX4ypSfGguFgC-IblTvO2ivM?usp=sharing) |\n| Inference Chat UI | LLaMa 3.2 Vision | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FUnsloth_Studio.ipynb) |\n| Conversational | LLaMa 3.2 (1B and 3B) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLlama3.2_(1B_and_3B)-Conversational.ipynb) |\n| ChatML | Mistral (7B) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F15F1xyn8497_dUbxZP4zWmPZ3PJx1Oymv?usp=sharing) |\n| Text Completion | Mistral (7B) | [![Open In Colab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FMistral_(7B)-Text_Completion.ipynb) |\n\n### Other Notebooks\n| Model | Type | Notebook Link |\n| --- | --- | --- |\n| **CodeForces cot Finetune\u003Cbr>for Reasoning on CodeForces** | Reasoning | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FCodeForces-cot-Finetune_for_Reasoning_on_CodeForces.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Synthetic Data Hackathon** | Synthetic Data | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FSynthetic_Data_Hackathon.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n| **Unsloth** | Studio | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FUnsloth_Studio.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"Open In Colab\">\u003C\u002Fa> |\n\n# 📒 Kaggle Notebooks\n\u003Cdetails>\n \u003Csummary>\n Click for all our Kaggle notebooks categorized by model:\n \u003C\u002Fsummary>\n\n### GRPO & Reinforcement Learning Notebooks\n| Model | Type | Notebook Link |\n| --- | --- | --- |\n| **Llama3.1** **(8B)** | GSM8K Math + vLLM | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Llama3.1_(8B)-GRPO.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **gpt oss** **(20B)** | Minesweeper Game | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-gpt_oss_(20B)_Reinforcement_Learning_GRPO_Minesweeper_Game_BF16.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **gpt oss** **(20B)** | Auto Kernel Creation | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-gpt_oss_(20B)_GRPO_BF16.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Qwen3** **(8B)** | DAPO Math + vLLM | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen3_8B_FP8_GRPO.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Llama3** **(8B)** | ORPO | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Llama3_(8B)-ORPO.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Qwen2.5 VL** **(7B)** | Vision Math + vLLM | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen2_5_7B_VL_GRPO.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Ministral3** **(3B)** | Sudoku | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Ministral_3_(3B)_Reinforcement_Learning_Sudoku_Game.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Zephyr** **(7B)** | DPO | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Zephyr_(7B)-DPO.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **gpt oss** **(20B)** | Auto Kernel Creation | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-gpt-oss-(20B)-GRPO.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **(A100) gpt oss** **(20B)** | Auto Kernel Creation | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-gpt-oss-(20B)_A100-GRPO.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Qwen2.5** **(3B)** | GSM8K Math + vLLM | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen2.5_(3B)-GRPO.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Llama3.2** **(1B)** | DAPO Math + vLLM | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Llama_FP8_GRPO.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Qwen3 VL** **(8B)** | Vision Math | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen3_VL_(8B)-Vision-GRPO.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Mistral v0.3** **(7B)** | GSM8K Math + vLLM | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Mistral_v0.3_(7B)-GRPO.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Gemma3** **(4B)** | Vision Math | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Gemma3_(4B)-Vision-GRPO.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Phi 4** **(14B)** | GSM8K Math + vLLM | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Phi_4_(14B)-GRPO.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Gemma3** **(1B)** | GSM8K Math | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Gemma3_(1B)-GRPO.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **DeepSeek R1 0528 Qwen3** **(8B)** | DAPO Math + vLLM | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-DeepSeek_R1_0528_Qwen3_(8B)_GRPO.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Qwen3** **(4B)** | DAPO Math + vLLM | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen3_(4B)-GRPO.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n\n### Tool Calling Notebooks\n| Model | Type | Notebook Link |\n| --- | --- | --- |\n| **Qwen2.5 Coder** **(1.5B)** | Tool Calling | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen2.5_Coder_(1.5B)-Tool_Calling.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n\n### Text-to-Speech (TTS) Notebooks\n| Model | Type | Notebook Link |\n| --- | --- | --- |\n| **Spark TTS** **(0.5B)** | TTS | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Spark_TTS_(0_5B).ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Orpheus** **(3B)** | TTS | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Orpheus_(3B)-TTS.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Llasa TTS** **(3B)** | TTS | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Llasa_TTS_(3B).ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Llasa TTS** **(1B)** | TTS | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Llasa_TTS_(1B).ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Sesame CSM** **(1B)** | TTS | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Sesame_CSM_(1B)-TTS.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Oute TTS** **(1B)** | TTS | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Oute_TTS_(1B).ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n\n### Vision (Multimodal) Notebooks\n| Model | Type | Notebook Link |\n| --- | --- | --- |\n| **Qwen2.5 VL** **(7B)** | Vision Math + vLLM (GRPO RL) | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen2_5_7B_VL_GRPO.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Qwen2.5 VL** **(7B)** | Vision | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen2.5_VL_(7B)-Vision.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **ERNIE 4 5 VL 28B A3B PT** | Vision | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-ERNIE_4_5_VL_28B_A3B_PT_Vision.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Qwen3 VL** **(8B)** | Vision | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen3_VL_(8B)-Vision.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Qwen3 VL** **(8B)** | Vision Math (GRPO RL) | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen3_VL_(8B)-Vision-GRPO.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Gemma3** **(4B)** | Vision Math (GRPO RL) | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Gemma3_(4B)-Vision-GRPO.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Qwen2 VL** **(7B)** | Vision | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen2_VL_(7B)-Vision.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Llama3.2** **(11B)** | Vision | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Llama3.2_(11B)-Vision.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Pixtral** **(12B)** | Vision | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Pixtral_(12B)-Vision.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Ministral3 VL** **(3B)** | Vision | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Ministral_3_VL_(3B)_Vision.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Gemma3** **(4B)** | Vision | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Gemma3_(4B)-Vision.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Gemma3N** **(4B)** | Vision | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Gemma3N_(4B)-Vision.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n\n### Embedding Notebooks\n| Model | Type | Notebook Link |\n| --- | --- | --- |\n| **ModernBert** | | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-ModernBert.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Qwen3 Embedding** **(4B)** | | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen3_Embedding_(4B).ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Qwen3 Embedding** **(0 6B)** | | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen3_Embedding_(0_6B).ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **EmbeddingGemma** **(300M)** | | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-EmbeddingGemma_(300M).ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **BGE M3** | | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-BGE_M3.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **ModernBERT** **(Large)** | Classification | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-bert_classification.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **All MiniLM L6 v2** | | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-All_MiniLM_L6_v2.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n\n### Speech-to-Text (STT) Notebooks\n| Model | Type | Notebook Link |\n| --- | --- | --- |\n| **Whisper** **(Large)** | Fine Tuning | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Whisper.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n\n### OCR Notebooks\n| Model | Type | Notebook Link |\n| --- | --- | --- |\n| **Deepseek OCR** **(3B)** | Fine Tuning | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Deepseek_OCR_(3B).ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Deepseek OCR** **(3B)** | Evaluation | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Deepseek_OCR_(3B)-Evaluation.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Deepseek OCR** **(3B)** | Eval | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Deepseek_OCR_(3B)-Eval.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Deepseek OCR 2** **(3B)** | | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Deepseek_OCR_2_(3B).ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Paddle OCR** **(1B)** | Vision | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Paddle_OCR_(1B)_Vision.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n\n### BERT Notebooks\n| Model | Type | Notebook Link |\n| --- | --- | --- |\n| **ModernBert** | | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-ModernBert.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **ModernBERT** **(Large)** | Classification | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-bert_classification.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n\n### Deepseek Notebooks\n| Model | Type | Notebook Link |\n| --- | --- | --- |\n| **Deepseek OCR** **(3B)** | Fine Tuning | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Deepseek_OCR_(3B).ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Deepseek OCR** **(3B)** | Evaluation | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Deepseek_OCR_(3B)-Evaluation.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Deepseek OCR** **(3B)** | Eval | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Deepseek_OCR_(3B)-Eval.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Deepseek OCR 2** **(3B)** | | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Deepseek_OCR_2_(3B).ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n\n### ERNIE Notebooks\n| Model | Type | Notebook Link |\n| --- | --- | --- |\n| **ERNIE 4 5 VL 28B A3B PT** | Vision | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-ERNIE_4_5_VL_28B_A3B_PT_Vision.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **ERNIE 4 5 21B A3B PT** | Conversational | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-ERNIE_4_5_21B_A3B_PT-Conversational.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n\n### GPT-OSS Notebooks\n| Model | Type | Notebook Link |\n| --- | --- | --- |\n| **gpt oss MXFP4** **(20B)** | Inference | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-GPT_OSS_MXFP4_(20B)-Inference.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **gpt oss** **(20B)** | Fine Tuning | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-gpt_oss_(20B)_500K_Context_Fine_tuning.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **gpt oss** **(20B)** | Fine Tuning | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-gpt-oss-(20B)-Fine-tuning.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **gpt oss BNB** **(20B)** | Inference | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-GPT_OSS_BNB_(20B)-Inference.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **(A100) gpt oss** **(120B)** | Fine Tuning | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-gpt-oss-(120B)_A100-Fine-tuning.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n\n### Gemma Notebooks\n| Model | Type | Notebook Link |\n| --- | --- | --- |\n| **Gemma3** **(4B)** | Conversational | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Gemma3_(4B).ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **(A100) Gemma3** **(27B)** | Conversational | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Gemma3_(27B)_A100-Conversational.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Gemma3** **(270M)** | Conversational | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Gemma3_(270M).ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Gemma3N** **(4B)** | Multimodal | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Gemma3N_(4B)-Conversational.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Gemma3N** **(4B)** | Audio | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Gemma3N_(4B)-Audio.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Gemma3N** **(2B)** | Inference | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Gemma3N_(2B)-Inference.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Gemma3** **(4B)** | Vision | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Gemma3_(4B)-Vision.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Gemma3N** **(4B)** | Vision | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Gemma3N_(4B)-Vision.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **EmbeddingGemma** **(300M)** | | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-EmbeddingGemma_(300M).ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Gemma2** **(9B)** | Alpaca | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Gemma2_(9B)-Alpaca.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Gemma2** **(2B)** | Alpaca | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Gemma2_(2B)-Alpaca.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **CodeGemma** **(7B)** | Conversational | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-CodeGemma_(7B)-Conversational.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n\n### Granite Notebooks\n| Model | Type | Notebook Link |\n| --- | --- | --- |\n| **Granite4.0** **(3B)** | Conversational | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Granite4.0.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Granite4.0** **(350M)** | Conversational | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Granite4.0_350M.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n\n### Hybrid Attention Notebooks\n| Model | Type | Notebook Link |\n| --- | --- | --- |\n| **Liquid LFM2** **(1.2B)** | Conversational | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Liquid_LFM2_(1.2B)-Conversational.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Falcon H1** **(0.5B)** | Alpaca | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Falcon_H1_(0.5B)-Alpaca.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n\n### Llama Notebooks\n| Model | Type | Notebook Link |\n| --- | --- | --- |\n| **Llama3.1** **(8B)** | Inference | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Llama3.1_(8B)-Inference.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Llama3** **(8B)** | Ollama | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Llama3_(8B)-Ollama.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Llama3** **(8B)** | Alpaca | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Llama3_(8B)-Alpaca.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Llama3.2** **(1B and 3B)** | Conversational | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Llama3.2_(1B_and_3B)-Conversational.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Llama3.2** **(1B)** | RAFT | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Llama3.2_(1B)-RAFT.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Llama3.1** **(8B)** | Alpaca | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Llama3.1_(8B)-Alpaca.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **(A100) Llama3.3** **(70B)** | Conversational | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Llama3.3_(70B)_A100-Conversational.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Llama3.2** **(11B)** | Vision | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Llama3.2_(11B)-Vision.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Llama3** **(8B)** | Conversational | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Llama3_(8B)-Conversational.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **TinyLlama** **(1.1B)** | Alpaca | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-TinyLlama_(1.1B)-Alpaca.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n\n### Mistral Notebooks\n| Model | Type | Notebook Link |\n| --- | --- | --- |\n| **Mistral v0.3** **(7B)** | Conversational | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Mistral_v0.3_(7B)-Conversational.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Magistral** **(24B)** | Reasoning Conversational | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Magistral_(24B)-Reasoning-Conversational.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Pixtral** **(12B)** | Vision | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Pixtral_(12B)-Vision.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Mistral Small** **(22B)** | Alpaca | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Mistral_Small_(22B)-Alpaca.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Ministral3 VL** **(3B)** | Vision | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Ministral_3_VL_(3B)_Vision.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Mistral v0.3** **(7B)** | Alpaca | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Mistral_v0.3_(7B)-Alpaca.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Mistral Nemo** **(12B)** | Alpaca | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Mistral_Nemo_(12B)-Alpaca.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n\n### Nemotron Notebooks\n| Model | Type | Notebook Link |\n| --- | --- | --- |\n| **(A100) Nemotron Nano 3 30B A3B** | | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Nemotron-Nano-3-30B-A3B_A100.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **(A100) Nemotron 3 Nano 30B A3B** | | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Nemotron-3-Nano-30B-A3B_A100.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n\n### Paddle Notebooks\n| Model | Type | Notebook Link |\n| --- | --- | --- |\n| **Paddle OCR** **(1B)** | Vision | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Paddle_OCR_(1B)_Vision.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n\n### Phi Notebooks\n| Model | Type | Notebook Link |\n| --- | --- | --- |\n| **Phi 4** | Conversational | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Phi_4-Conversational.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Phi 3.5 Mini** | Conversational | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Phi_3.5_Mini-Conversational.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Phi 3 Medium** | Conversational | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Phi_3_Medium-Conversational.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n\n### Qwen Notebooks\n| Model | Type | Notebook Link |\n| --- | --- | --- |\n| **Qwen3** **(4B)** | QAT | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen3_(4B)_Instruct-QAT.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Qwen3** **(4B)** | Conversational | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen3_(4B)-Instruct.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Qwen2.5 VL** **(7B)** | Vision | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen2.5_VL_(7B)-Vision.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Qwen3 VL** **(8B)** | Vision | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen3_VL_(8B)-Vision.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **(A100) Qwen3** **(32B)** | Reasoning Conversational | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen3_(32B)_A100-Reasoning-Conversational.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Qwen3** **(14B)** | Reasoning Conversational | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen3_(14B)-Reasoning-Conversational.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Qwen3** **(14B)** | | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen3_(14B).ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Qwen2 VL** **(7B)** | Vision | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen2_VL_(7B)-Vision.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Qwen3** **(4B)** | Thinking | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen3_(4B)-Thinking.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Qwen2.5** **(7B)** | Alpaca | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen2.5_(7B)-Alpaca.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Qwen2.5 Coder** **(14B)** | Conversational | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen2.5_Coder_(14B)-Conversational.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Qwen3 Embedding** **(4B)** | | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen3_Embedding_(4B).ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Qwen3 Embedding** **(0 6B)** | | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen3_Embedding_(0_6B).ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Qwen3** **(14B)** | Alpaca | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen3_(14B)-Alpaca.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Qwen2** **(7B)** | Alpaca | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen2_(7B)-Alpaca.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n\n### Text Completion \u002F Continued Pretraining Notebooks\n| Model | Type | Notebook Link |\n| --- | --- | --- |\n| **Mistral v0.3** **(7B)** | CPT | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Mistral_v0.3_(7B)-CPT.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Mistral** **(7B)** | Text Completion | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Mistral_(7B)-Text_Completion.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n\n### Other Notebooks\n| Model | Type | Notebook Link |\n| --- | --- | --- |\n| **CodeForces cot Finetune\u003Cbr>for Reasoning on CodeForces** | Reasoning | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-CodeForces-cot-Finetune_for_Reasoning_on_CodeForces.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n| **Unsloth** | Studio | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Unsloth_Studio.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"Open in Kaggle\">\u003C\u002Fa> |\n\n\u003C\u002Fdetails>\n\n\n\u003C!-- End of Notebook Links -->\n\n## Known Issues \u002F Environment Notes\n\n- **NumPy 2.x ↔ soxr**: NumPy 2.x breaks soxr, causing Unsloth import failures. Pin `numpy\u003C2` to resolve. Use `pip install --force-reinstall \"numpy\u003C2\"` if needed. _Impact: Prevents Unsloth from running._\n- **soxr reinstall**: `pip install --force-reinstall soxr` can pull NumPy 2.x back unless using `--no-deps`. Use `pip install --force-reinstall --no-deps soxr` to avoid this. _Impact: May reintroduce NumPy 2.x and break Unsloth imports._\n- **typing_extensions**: Older typing_extensions can break torch import (TypeIs missing) until upgraded. Upgrade with `pip install --upgrade typing_extensions`. _Impact: Prevents PyTorch from importing correctly._\n- **Resolver warnings**: Pinning `numpy\u003C2` can cause pip resolver warnings with SciPy\u002FNumba; typically non-fatal. _Impact: Cosmetic warnings only, does not affect functionality._\n- **ROCm \u002F triton_key**: LoRA backward can crash under `torch.compile` if Triton lacks `triton_key`; workaround is to disable Inductor\u002Fcompile on ROCm (handled in code now, but worth noting). _Impact: May cause training crashes on AMD GPUs when using torch.compile._\n\n# ✨ Contributing to Notebooks\n\nIf you'd like to contribute to our notebooks, here's a guide to get you started:\n\n1. **Find the Template:** We've provided a template notebook called `Template_Notebook.ipynb` in the root directory of this project. This template contains the basic structure and formatting guidelines for all notebooks in this collection.\n2. **Create Your Notebook:**\n * Make a copy of `Template_Notebook.ipynb`.\n * Rename the copied file to follow this naming convention:\n * **LLM Notebooks:** `\u003CModel Name>-\u003CType>.ipynb` (e.g., `Mistral_v0.3_(7B)-Alpaca.ipynb`)\n * **Vision Notebooks:** `\u003CModel Name>-Vision.ipynb` (e.g., `Llava_v1.6_(7B)-Vision.ipynb`)\n * **Example of `\u003CType>`:** `Alpaca`, `Conversational`, `CPT`, `DPO`, `ORPO`, `Text_Completion`, `CSV`, `Inference`, `Unsloth_Studio`\n \u003C!-- * Modify the content of your notebook, adding your code, explanations, and any other relevant information. Make sure to follow the structure and guidelines from the template. -->\n3. **Place in `original_template`:** Once your notebook is ready, move it to the `original_template` directory.\n4. **Update Notebooks:** Run the following command in your terminal:\n ```bash\n python update_all_notebooks.py\n ```\n This script will automatically:\n * Copy your notebook from `original_template` to the `notebooks` directory.\n * Update the notebook's internal sections (like Installation, News) to ensure consistency.\n * Add your notebook to the appropriate list in this `README.md` file.\n5. **Create a Pull Request:** After that, just create a pull request (PR) to merge your changes, making it available for everyone!\n * We appreciate your contributions and look forward to reviewing your notebooks!\n","\u003Cdiv align=\"center\">\n\n \u003Ca href=\"https:\u002F\u002Funsloth.ai\">\u003Cpicture>\n \u003Csource media=\"(prefers-color-scheme: dark)\" srcset=\"https:\u002F\u002Fraw.githubusercontent.com\u002Funslothai\u002Funsloth\u002Fmain\u002Fimages\u002Funsloth%20logo%20white%20text.png\">\n \u003Csource media=\"(prefers-color-scheme: light)\" srcset=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Funslothai_notebooks_readme_f98b20d7f469.png\">\n \u003Cimg alt=\"unsloth logo\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Funslothai_notebooks_readme_f98b20d7f469.png\" height=\"110\" style=\"max-width: 100%;\">\n \u003C\u002Fpicture>\u003C\u002Fa>\n \n\u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002Fgpt-oss-(20B)-Fine-tuning.ipynb\">\u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Funslothai_notebooks_readme_6e935a75881d.png\" height=\"48\">\u003C\u002Fa>\n\u003Ca href=\"https:\u002F\u002Fdiscord.gg\u002Funsloth\">\u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Funslothai_notebooks_readme_a85808c43728.png\" height=\"48\">\u003C\u002Fa>\n\u003Ca href=\"https:\u002F\u002Funsloth.ai\u002Fdocs\u002Fget-started\u002Funsloth-notebooks\">\u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Funslothai_notebooks_readme_cb417d7bd7e4.png\" height=\"48\">\u003C\u002Fa>\n\n\u003C\u002Fdiv>\n\n## 📒 微调笔记本\n以下是按模型分类的 Colab 笔记本。你也可以在我们的文档中查看所有 [笔记本](https:\u002F\u002Funsloth.ai\u002Fdocs\u002Fget-started\u002Funsloth-notebooks)。\u003Cbr>这些笔记本可以在本地运行，包含数据准备、训练和推理等内容。请阅读我们的 [微调指南](https:\u002F\u002Funsloth.ai\u002Fdocs\u002Fget-started\u002Ffine-tuning-llms-guide)。\n\n### 主要笔记本\n| 模型 | 类型 | 笔记本链接 |\n|-----------------------------|----------------|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|\n| **Unsloth Studio** | 聊天界面 | [![在 Colab 中打开](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FUnsloth_Studio.ipynb) |\n| **Gemma 4 (E2B)** | 视觉 | [![在 Colab 中打开](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma4_%28E2B%29-Vision.ipynb) |\n| **Qwen3.5 (4B)** | 视觉 | [![在 Colab 中打开](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_5_%284B%29_Vision.ipynb) |\n| **Qwen3.5 (2B)** | 视觉 | [![在 Colab 中打开](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_5_%282B%29_Vision.ipynb) |\n| **gpt-oss (20B)** | 微调 | [![在 Colab 中打开](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002Fgpt-oss-%2820B%29-Fine-tuning.ipynb) |\n| **gpt-oss (20B)** | GRPO | [![在 Colab 中打开](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002Fgpt-oss-%2820B%29-GRPO.ipynb) |\n| **Qwen3 (14B)** | 对话 | [![在 Colab 中打开](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_%2814B%29-Reasoning-Conversational.ipynb) |\n| **Qwen3-VL (8B)** | 视觉 | [![在 Colab 中打开](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_VL_%288B%29-Vision.ipynb) |\n| **Qwen3-Embedding (0.6B)** | 嵌入 | [![在 Colab 中打开](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_Embedding_(0_6B).ipynb) |\n| **Qwen3: 高级 GRPO** | GRPO | [![在 Colab 中打开](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_%284B%29-GRPO.ipynb) |\n| **Gemma 3 (4B)** | 视觉 | [![在 Colab 中打开](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma3_%284B%29-Vision.ipynb) |\n| **Gemma 3N (4B)** | 音频 | [![在 Colab 中打开](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma3N_%284B%29-Audio.ipynb) |\n| **embeddinggemma (300M)** | 嵌入 | [![在 Colab 中打开](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FEmbeddingGemma_%28300M%29.ipynb) |\n| **Mistral Ministral 3 (3B)**| 视觉 | [![在 Colab 中打开](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FMinistral_3_VL_%283B%29_Vision.ipynb) |\n| **Mistral v0.3 (7B)** | 视觉 | [![在 Colab 中打开](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FMistral_v0.3_(7B)-Alpaca.ipynb) |\n| **Llama 3.1 (8B) Alpaca** | Alpaca | [![在 Colab 中打开](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLlama3.1_%288B%29-Alpaca.ipynb) |\n| **Llama 3.2 (1B + 3B)**| 对话 | [![在 Colab 中打开](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLlama3.2_%281B_and_3B%29-Conversational.ipynb) |\n| **Phi-4 (14B)** | 对话 | [![在 Colab 中打开](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FPhi_4-Conversational.ipynb) |\n| **Orpheus-TTS (3B)** | TTS | [![在 Colab 中打开](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FOrpheus_%283B%29-TTS.ipynb) |\n\n\u003C!-- 🛑 🚨 请勿手动编辑此部分，直到 `end of notebook links`!! 🛑 🚨 -->\n\u003C!-- 🛑 🚨 本部分由 `update_all_notebooks.py` 自动生成 🛑 🚨 -->\n\n\u003C!-- 编辑开始 -->\n\n### GRPO 与强化学习笔记本\n| 模型 | 类型 | 笔记本链接 |\n| --- | --- | --- |\n| **Llama3.1** **(8B)** | GSM8K 数学 + vLLM | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLlama3.1_(8B)-GRPO.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **NeMo Gym Sudoku** | 数独 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FNeMo-Gym-Sudoku.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **NeMo Gym Multi Environment** | 多环境 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FNeMo-Gym-Multi-Environment.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **gpt oss BF16** **(20B)** | 2048 游戏 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002Fgpt_oss_(20B)_Reinforcement_Learning_2048_Game_BF16.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **gpt oss** **(20B)** | 扫雷游戏 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002Fgpt_oss_(20B)_Reinforcement_Learning_GRPO_Minesweeper_Game_BF16.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **gpt oss** **(20B)** | 自动内核创建 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002Fgpt_oss_(20B)_GRPO_BF16.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Qwen3** **(8B)** | DAPO 数学 + vLLM | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_8B_FP8_GRPO.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Llama3** **(8B)** | ORPO | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLlama3_(8B)-ORPO.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Openenv wordle** | Wordle + vLLM | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FOpenenv_wordle_grpo.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Qwen2.5 VL** **(7B)** | 视觉数学 + vLLM | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen2_5_7B_VL_GRPO.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Zephyr** **(7B)** | DPO | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FZephyr_(7B)-DPO.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **(OpenEnv) gpt oss BF16** **(20B)** | 2048 游戏 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FOpenEnv_gpt_oss_(20B)_Reinforcement_Learning_2048_Game_BF16.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **gpt oss** **(20B)** | 自动内核创建 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002Fgpt-oss-(20B)-GRPO.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **gpt oss** **(20B)** | 2048 游戏 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002Fgpt_oss_(20B)_Reinforcement_Learning_2048_Game.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **(OpenEnv) gpt oss** **(20B)** | 2048 游戏 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FOpenEnv_gpt_oss_(20B)_Reinforcement_Learning_2048_Game.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **(DGX Spark) gpt oss** **(20B)** | 2048 游戏 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002Fgpt_oss_(20B)_Reinforcement_Learning_2048_Game_DGX_Spark.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **(A100) gpt oss** **(20B)** | 自动内核创建 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002Fgpt-oss-(20B)_A100-GRPO.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Qwen2.5** **(3B)** | GSM8K 数学 + vLLM | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen2.5_(3B)-GRPO.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Llama3.2** **(1B)** | DAPO 数学 + vLLM | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLlama_FP8_GRPO.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Qwen3 VL** **(8B)** | 视觉数学 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_VL_(8B)-Vision-GRPO.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Mistral v0.3** **(7B)** | GSM8K 数学 + vLLM | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FMistral_v0.3_(7B)-GRPO.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Qwen3 5** **(4B)** | 视觉数学 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_5_(4B)_Vision_GRPO.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Gemma3** **(4B)** | 视觉数学 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma3_(4B)-Vision-GRPO.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Phi 4** **(14B)** | GSM8K 数学 + vLLM | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FPhi_4_(14B)-GRPO.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Gemma3** **(1B)** | GSM8K 数学 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma3_(1B)-GRPO.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **DeepSeek R1 0528 Qwen3** **(8B)** | DAPO 数学 + vLLM | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FDeepSeek_R1_0528_Qwen3_(8B)_GRPO.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **LFM2.5** **(1.2B)** | DAPO 数学 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLFM2.5_(1.2B)-GRPO.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Qwen3** **(4B)** | DAPO 数学 + vLLM | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_(4B)-GRPO.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Ministral3** **(3B)** | 数独 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FMinistral_3_(3B)_Reinforcement_Learning_Sudoku_Game.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n\n### 工具调用笔记本\n| 模型 | 类型 | 笔记本链接 |\n| --- | --- | --- |\n| **Qwen2.5 Coder** **(1.5B)** | 工具调用 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen2.5_Coder_(1.5B)-Tool_Calling.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **FunctionGemma** **(270M)** | 工具调用 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FFunctionGemma_(270M)-Multi-Turn-Tool-Calling.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **FunctionGemma** **(270M)** | 移动操作 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FFunctionGemma_(270M)-Mobile-Actions.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **FunctionGemma** **(270M)** | 推理 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FFunctionGemma_(270M)-LMStudio.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **FunctionGemma** **(270M)** | 对话 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FFunctionGemma_(270M).ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n\n### 文本转语音（TTS）笔记本\n| 模型 | 类型 | 笔记本链接 |\n| --- | --- | --- |\n| **Spark TTS** **(0.5B)** | TTS | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FSpark_TTS_(0_5B).ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **Orpheus** **(3B)** | TTS | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FOrpheus_(3B)-TTS.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **Llasa TTS** **(3B)** | TTS | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLlasa_TTS_(3B).ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **Llasa TTS** **(1B)** | TTS | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLlasa_TTS_(1B).ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **Sesame CSM** **(1B)** | TTS | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FSesame_CSM_(1B)-TTS.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **Oute TTS** **(1B)** | TTS | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FOute_TTS_(1B).ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n\n### 视觉（多模态）笔记本\n| 模型 | 类型 | 笔记本链接 |\n| --- | --- | --- |\n| **Qwen2.5 VL** **(7B)** | 视觉数学 + vLLM (GRPO RL) | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen2_5_7B_VL_GRPO.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **Gemma4** **(E4B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma4_(E4B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **Gemma4** **(E2B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma4_(E2B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **Gemma4** **(26B A4B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma4_(26B_A4B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **Gemma4** **(31B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma4_(31B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **Qwen2.5 VL** **(7B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen2.5_VL_(7B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **ERNIE 4 5 VL 28B A3B PT** | 视觉 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FERNIE_4_5_VL_28B_A3B_PT_Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **Qwen3 VL** **(8B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_VL_(8B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **Qwen3 VL** **(8B)** | 视觉数学 (GRPO RL) | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_VL_(8B)-Vision-GRPO.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **Qwen3 5** **(4B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_5_(4B)_Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **Qwen3 5** **(4B)** | 视觉数学 (GRPO RL) | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_5_(4B)_Vision_GRPO.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **Gemma3** **(4B)** | 视觉数学 (GRPO RL) | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma3_(4B)-Vision-GRPO.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **Qwen3 5** **(0.8B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_5_(0.8B)_Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **Qwen2 VL** **(7B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen2_VL_(7B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **Llama3.2** **(11B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLlama3.2_(11B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **Qwen3 5** **(2B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_5_(2B)_Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **Pixtral** **(12B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FPixtral_(12B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **LFM2.5 VL** **(1.6B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLFM2.5_VL_(1.6B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **Ministral3 VL** **(3B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FMinistral_3_VL_(3B)_Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **Gemma3** **(4B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma3_(4B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **Gemma3N** **(4B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma3N_(4B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n\n### 嵌入笔记本\n| 模型 | 类型 | 笔记本链接 |\n| --- | --- | --- |\n| **ModernBert** | | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FModernBert.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **Qwen3 Embedding** **(4B)** | | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_Embedding_(4B).ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **Qwen3 Embedding** **(0.6B)** | | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_Embedding_(0.6B).ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **EmbeddingGemma** **(300M)** | | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FEmbeddingGemma_(300M).ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **BGE M3** | | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FBGE_M3.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **ModernBERT** **(Large)** | 分类 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002Fbert_classification.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **All MiniLM L6 v2** | | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FAll_MiniLM_L6_v2.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n\n### 语音转文本（STT）笔记本\n| 模型 | 类型 | 笔记本链接 |\n| --- | --- | --- |\n| **Whisper** **(Large)** | 微调 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FWhisper.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n\n### OCR笔记本\n| 模型 | 类型 | 笔记本链接 |\n| --- | --- | --- |\n| **Deepseek OCR** **(3B)** | 微调 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FDeepseek_OCR_(3B).ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **Deepseek OCR** **(3B)** | 评估 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FDeepseek_OCR_(3B)-Evaluation.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **Deepseek OCR** **(3B)** | 测试 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FDeepseek_OCR_(3B)-Eval.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **Deepseek OCR 2** **(3B)** | | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FDeepseek_OCR_2_(3B).ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **Paddle OCR** **(1B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FPaddle_OCR_(1B)_Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n\n### BERT笔记本\n| 模型 | 类型 | 笔记本链接 |\n| --- | --- | --- |\n| **ModernBert** | | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FModernBert.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **ModernBERT** **(Large)** | 分类 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002Fbert_classification.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n\n### Deepseek笔记本\n| 模型 | 类型 | 笔记本链接 |\n| --- | --- | --- |\n| **Deepseek OCR** **(3B)** | 微调 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FDeepseek_OCR_(3B).ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **Deepseek OCR** **(3B)** | 评估 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FDeepseek_OCR_(3B)-Evaluation.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **Deepseek OCR** **(3B)** | 测试 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FDeepseek_OCR_(3B)-Eval.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **Deepseek OCR 2** **(3B)** | | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FDeepseek_OCR_2_(3B).ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n\n### ERNIE笔记本\n| 模型 | 类型 | 笔记本链接 |\n| --- | --- | --- |\n| **ERNIE 4.5 VL 28B A3B PT** | 视觉 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FERNIE_4_5_VL_28B_A3B_PT_Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **ERNIE 4.5 21B A3B PT** | 对话 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FERNIE_4_5_21B_A3B_PT-Conversational.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n\n### GLM笔记本\n| 模型 | 类型 | 笔记本链接 |\n| --- | --- | --- |\n| **(A100) GLM Flash(80GB)** | | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGLM_Flash_A100(80GB).ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n\n### GPT-OSS 笔记本\n| 模型 | 类型 | 笔记本链接 |\n| --- | --- | --- |\n| **gpt oss MXFP4** **(20B)** | 推理 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGPT_OSS_MXFP4_(20B)-Inference.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **gpt oss** **(20B)** | 微调 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002Fgpt_oss_(20B)_500K_Context_Fine_tuning.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **gpt oss** **(20B)** | 微调 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002Fgpt-oss-(20B)-Fine-tuning.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **gpt oss BNB** **(20B)** | 推理 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGPT_OSS_BNB_(20B)-Inference.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **(A100) gpt oss** **(120B)** | 微调 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002Fgpt-oss-(120B)_A100-Fine-tuning.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n\n### Gemma 4 笔记本\n| 模型 | 类型 | 笔记本链接 |\n| --- | --- | --- |\n| **Gemma4** **(E4B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma4_(E4B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Gemma4** **(E4B)** | 对话 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma4_(E4B)-Text.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Gemma4** **(E4B)** | 音频 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma4_(E4B)-Audio.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Gemma4** **(E2B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma4_(E2B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Gemma4** **(E2B)** | 音频 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma4_(E2B)-Audio.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Gemma4** **(E2B)** | 对话 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma4_(E2B)-Text.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Gemma4** **(26B A4B)** | 对话 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma4_(26B_A4B)-Text.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Gemma4** **(26B A4B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma4_(26B_A4B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Gemma4** **(31B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma4_(31B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Gemma4** **(31B)** | 对话 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma4_(31B)-Text.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n\n### Gemma 笔记本\n| 模型 | 类型 | 笔记本链接 |\n| --- | --- | --- |\n| **Gemma3** **(4B)** | 对话 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma3_(4B).ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **(A100) Gemma3** **(27B)** | 对话 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma3_(27B)_A100-Conversational.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Gemma3** **(270M)** | 对话 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma3_(270M).ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Gemma3** **(270M)** | | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma3_(270M)_Phone_Deployment.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Gemma3N** **(4B)** | 多模态 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma3N_(4B)-Conversational.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Gemma3N** **(4B)** | 音频 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma3N_(4B)-Audio.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Gemma3N** **(2B)** | 推理 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma3N_(2B)-Inference.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Gemma3** **(4B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma3_(4B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Gemma3N** **(4B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma3N_(4B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **EmbeddingGemma** **(300M)** | | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FEmbeddingGemma_(300M).ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Gemma2** **(9B)** | Alpaca | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma2_(9B)-Alpaca.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Gemma2** **(2B)** | Alpaca | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGemma2_(2B)-Alpaca.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **CodeGemma** **(7B)** | 对话 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FCodeGemma_(7B)-Conversational.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n\n### Granite 笔记本\n| 模型 | 类型 | 笔记本链接 |\n| --- | --- | --- |\n| **Granite4.0** **(3B)** | 对话 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGranite4.0.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Granite4.0** **(350M)** | 对话 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FGranite4.0_350M.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n\n### 混合注意力笔记本\n| 模型 | 类型 | 笔记本链接 |\n| --- | --- | --- |\n| **LFM2.5** **(1.2B)** | 对话 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLFM2.5_(1.2B)-Conversational.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **LFM2.5 VL** **(1.6B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLFM2.5_VL_(1.6B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Liquid LFM2** **(1.2B)** | 对话 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLiquid_LFM2_(1.2B)-Conversational.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Liquid LFM2** | 对话 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLiquid_LFM2-Conversational.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **LFM2.5** **(1.2B)** | | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLFM2.5_(1.2B)-Translation.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Falcon H1** **(0.5B)** | Alpaca | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FFalcon_H1_(0.5B)-Alpaca.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Falcon H1** | Alpaca | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FFalcon_H1-Alpaca.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n\n### Llama 笔记本\n| 模型 | 类型 | 笔记本链接 |\n| --- | --- | --- |\n| **Llama3.1** **(8B)** | 推理 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLlama3.1_(8B)-Inference.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Llama3** **(8B)** | Ollama | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLlama3_(8B)-Ollama.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Llama3** **(8B)** | Alpaca | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLlama3_(8B)-Alpaca.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Llama3.2** **(1B 和 3B)** | 对话 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLlama3.2_(1B_and_3B)-Conversational.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Llama3.2** **(1B)** | RAFT | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLlama3.2_(1B)-RAFT.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Llama3.1** **(8B)** | Alpaca | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLlama3.1_(8B)-Alpaca.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **(A100) Llama3.3** **(70B)** | 对话 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLlama3.3_(70B)_A100-Conversational.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Llama3.2** **(11B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLlama3.2_(11B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Llama3** **(8B)** | 对话 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLlama3_(8B)-Conversational.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **TinyLlama** **(1.1B)** | Alpaca | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FTinyLlama_(1.1B)-Alpaca.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n\n### Mistral 笔记本\n| 模型 | 类型 | 笔记本链接 |\n| --- | --- | --- |\n| **Mistral v0.3** **(7B)** | 对话 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FMistral_v0.3_(7B)-Conversational.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Magistral** **(24B)** | 理性对话 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FMagistral_(24B)-Reasoning-Conversational.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Pixtral** **(12B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FPixtral_(12B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Mistral Small** **(22B)** | Alpaca | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FMistral_Small_(22B)-Alpaca.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Ministral3 VL** **(3B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FMinistral_3_VL_(3B)_Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Mistral v0.3** **(7B)** | Alpaca | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FMistral_v0.3_(7B)-Alpaca.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Mistral Nemo** **(12B)** | Alpaca | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FMistral_Nemo_(12B)-Alpaca.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n\n### Nemotron 笔记本\n| 模型 | 类型 | 笔记本链接 |\n| --- | --- | --- |\n| **(A100) Nemotron Nano 3 30B A3B** | | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FNemotron-Nano-3-30B-A3B_A100.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **(A100) Nemotron 3 Nano 30B A3B** | | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FNemotron-3-Nano-30B-A3B_A100.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n\n### Paddle 笔记本\n| 模型 | 类型 | 笔记本链接 |\n| --- | --- | --- |\n| **Paddle OCR** **(1B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FPaddle_OCR_(1B)_Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n\n### Phi 笔记本\n| 型号 | 类型 | 笔记本链接 |\n| --- | --- | --- |\n| **Phi 4** | 对话型 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FPhi_4-Conversational.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Phi 3.5 Mini** | 对话型 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FPhi_3.5_Mini-Conversational.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Phi 3 Medium** | 对话型 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FPhi_3_Medium-Conversational.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n\n### 通义千问笔记本\n| 模型 | 类型 | 笔记本链接 |\n| --- | --- | --- |\n| **Qwen3** **(0.6B)** | 推理对话 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_(0.6B)-Reasoning-Conversational-ExecuTorch.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Qwen3** **(0.6B)** | | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_(0.6B)-Phone_Deployment.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Qwen3** **(4B)** | QAT | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_(4B)_Instruct-QAT.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Qwen3** **(4B)** | 对话 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_(4B)-Instruct.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Qwen2.5 VL** **(7B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen2.5_VL_(7B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Qwen3 VL** **(8B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_VL_(8B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Qwen3.5 MoE** | | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3.5_MoE.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **(A100) Qwen3.5 27B(80GB)** | | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3.5_27B_A100(80GB).ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **(A100) Qwen3** **(32B)** | 推理对话 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_(32B)_A100-Reasoning-Conversational.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Qwen3.5** **(4B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3.5_(4B)_Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Qwen3** **(14B)** | 推理对话 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_(14B)-Reasoning-Conversational.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Qwen3** **(14B)** | | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_(14B).ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Qwen3.5** **(0.8B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3.5_(0.8B)_Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Qwen2 VL** **(7B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen2_VL_(7B)-Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Qwen3** **(4B)** | 思考 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_(4B)-Thinking.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Qwen3 MoE** | | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_MoE.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Qwen3.5** **(2B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3.5_(2B)_Vision.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Qwen2.5** **(7B)** | 阿尔帕卡 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen2.5_(7B)-Alpaca.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Qwen2.5 Coder** **(14B)** | 对话 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen2.5_Coder_(14B)-Conversational.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Qwen3 Embedding** **(4B)** | | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_Embedding_(4B).ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Qwen3 Embedding** **(0.6B)** | | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_Embedding_(0.6B).ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Qwen3** **(14B)** | 阿尔帕卡 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen3_(14B)-Alpaca.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **Qwen2** **(7B)** | 阿尔帕卡 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen2_(7B)-Alpaca.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n| **TinyQwen3 MoE** | | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FTinyQwen3_MoE.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在 Colab 中打开\">\u003C\u002Fa> |\n\n### 文本补全 \u002F 继续预训练笔记本\n| 模型 | 类型 | 笔记本链接 |\n| --- | --- | --- |\n| **LFM2.5** **(1.2B)** | 文本补全 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLFM2.5_(1.2B)-Text_Completion.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **Mistral v0.3** **(7B)** | CPT | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FMistral_v0.3_(7B)-CPT.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **Mistral** **(7B)** | 文本补全 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FMistral_(7B)-Text_Completion.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n\n### 特定用例笔记本\n| 用例 | 模型 | 笔记本链接 |\n| --- | --- | --- |\n| 文本分类 | Llama 3.1 (8B) | [![在Colab中打开](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ftimothelaborie\u002Ftext_classification_scripts\u002Fblob\u002Fmain\u002Funsloth_classification.ipynb) |\n| 工具调用 | Qwen2.5-Coder (1.5B) | [![在Colab中打开](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FQwen2.5_Coder_(1.5B)-Tool_Calling.ipynb) |\n| 多数据集 | | [![在Colab中打开](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1njCCbE1YVal9xC83hjdo2hiGItpY_D6t?usp=sharing) |\n| KTO | Qwen2.5-Instruct (1.5B) | [![在Colab中打开](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1MRgGtLWuZX4ypSfGguFgC-IblTvO2ivM?usp=sharing) |\n| 推理聊天界面 | LLaMa 3.2 Vision | [![在Colab中打开](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FUnsloth_Studio.ipynb) |\n| 对话式 | LLaMa 3.2 (1B 和 3B) | [![在Colab中打开](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FLlama3.2_(1B_and_3B)-Conversational.ipynb) |\n| ChatML | Mistral (7B) | [![在Colab中打开](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F15F1xyn8497_dUbxZP4zWmPZ3PJx1Oymv?usp=sharing) |\n| 文本补全 | Mistral (7B) | [![在Colab中打开](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FMistral_(7B)-Text_Completion.ipynb) |\n\n### 其他笔记本\n| 模型 | 类型 | 笔记本链接 |\n| --- | --- | --- |\n| **CodeForces cot 微调\u003Cbr>用于 CodeForces 上的推理** | 推理 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FCodeForces-cot-Finetune_for_Reasoning_on_CodeForces.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **合成数据黑客马拉松** | 合成数据 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FSynthetic_Data_Hackathon.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n| **Unsloth** | 工作室 | \u003Ca href=\"https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FUnsloth_Studio.ipynb\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg\" alt=\"在Colab中打开\">\u003C\u002Fa> |\n\n# 📒 Kaggle 笔记本\n\u003Cdetails>\n \u003Csummary>\n 点击查看我们按模型分类的所有 Kaggle 笔记本：\n \u003C\u002Fsummary>\n\n### GRPO 与强化学习笔记本\n| 模型 | 类型 | 笔记本链接 |\n| --- | --- | --- |\n| **Llama3.1** **(8B)** | GSM8K 数学 + vLLM | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Llama3.1_(8B)-GRPO.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n| **gpt oss** **(20B)** | 扫雷游戏 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-gpt_oss_(20B)_Reinforcement_Learning_GRPO_Minesweeper_Game_BF16.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n| **gpt oss** **(20B)** | 自动内核创建 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-gpt_oss_(20B)_GRPO_BF16.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n| **Qwen3** **(8B)** | DAPO 数学 + vLLM | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen3_8B_FP8_GRPO.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n| **Llama3** **(8B)** | ORPO | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Llama3_(8B)-ORPO.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n| **Qwen2.5 VL** **(7B)** | 视觉数学 + vLLM | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen2_5_7B_VL_GRPO.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n| **Ministral3** **(3B)** | 数独 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Ministral_3_(3B)_Reinforcement_Learning_Sudoku_Game.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n| **Zephyr** **(7B)** | DPO | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Zephyr_(7B)-DPO.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n| **gpt oss** **(20B)** | 自动内核创建 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-gpt-oss-(20B)-GRPO.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n| **(A100) gpt oss** **(20B)** | 自动内核创建 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-gpt-oss-(20B)_A100-GRPO.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n| **Qwen2.5** **(3B)** | GSM8K 数学 + vLLM | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen2.5_(3B)-GRPO.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n| **Llama3.2** **(1B)** | DAPO 数学 + vLLM | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Llama_FP8_GRPO.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n| **Qwen3 VL** **(8B)** | 视觉数学 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen3_VL_(8B)-Vision-GRPO.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n| **Mistral v0.3** **(7B)** | GSM8K 数学 + vLLM | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Mistral_v0.3_(7B)-GRPO.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n| **Gemma3** **(4B)** | 视觉数学 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Gemma3_(4B)-Vision-GRPO.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n| **Phi 4** **(14B)** | GSM8K 数学 + vLLM | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Phi_4_(14B)-GRPO.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n| **Gemma3** **(1B)** | GSM8K 数学 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Gemma3_(1B)-GRPO.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n| **DeepSeek R1 0528 Qwen3** **(8B)** | DAPO 数学 + vLLM | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-DeepSeek_R1_0528_Qwen3_(8B)_GRPO.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n| **Qwen3** **(4B)** | DAPO 数学 + vLLM | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen3_(4B)-GRPO.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n\n### 工具调用笔记本\n| 模型 | 类型 | 笔记本链接 |\n| --- | --- | --- |\n| **Qwen2.5 Coder** **(1.5B)** | 工具调用 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen2.5_Coder_(1.5B)-Tool_Calling.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n\n### 文本转语音（TTS）笔记本\n| 模型 | 类型 | 笔记本链接 |\n| --- | --- | --- |\n| **Spark TTS** **(0.5B)** | TTS | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Spark_TTS_(0_5B).ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **Orpheus** **(3B)** | TTS | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Orpheus_(3B)-TTS.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **Llasa TTS** **(3B)** | TTS | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Llasa_TTS_(3B).ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **Llasa TTS** **(1B)** | TTS | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Llasa_TTS_(1B).ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **Sesame CSM** **(1B)** | TTS | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Sesame_CSM_(1B)-TTS.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **Oute TTS** **(1B)** | TTS | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Oute_TTS_(1B).ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n\n### 视觉（多模态）笔记本\n| 模型 | 类型 | 笔记本链接 |\n| --- | --- | --- |\n| **Qwen2.5 VL** **(7B)** | 视觉数学 + vLLM (GRPO RL) | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen2_5_7B_VL_GRPO.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **Qwen2.5 VL** **(7B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen2.5_VL_(7B)-Vision.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **ERNIE 4 5 VL 28B A3B PT** | 视觉 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-ERNIE_4_5_VL_28B_A3B_PT_Vision.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **Qwen3 VL** **(8B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen3_VL_(8B)-Vision.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **Qwen3 VL** **(8B)** | 视觉数学 (GRPO RL) | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen3_VL_(8B)-Vision-GRPO.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **Gemma3** **(4B)** | 视觉数学 (GRPO RL) | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Gemma3_(4B)-Vision-GRPO.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **Qwen2 VL** **(7B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen2_VL_(7B)-Vision.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **Llama3.2** **(11B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Llama3.2_(11B)-Vision.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **Pixtral** **(12B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Pixtral_(12B)-Vision.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **Ministral3 VL** **(3B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Ministral_3_VL_(3B)_Vision.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **Gemma3** **(4B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Gemma3_(4B)-Vision.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **Gemma3N** **(4B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Gemma3N_(4B)-Vision.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n\n### 嵌入模型笔记本\n| 模型 | 类型 | 笔记本链接 |\n| --- | --- | --- |\n| **ModernBert** | | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-ModernBert.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **Qwen3 Embedding** **(4B)** | | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen3_Embedding_(4B).ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **Qwen3 Embedding** **(0.6B)** | | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen3_Embedding_(0.6B).ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **EmbeddingGemma** **(300M)** | | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-EmbeddingGemma_(300M).ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **BGE M3** | | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-BGE_M3.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **ModernBERT** **(Large)** | 分类 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-bert_classification.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **All MiniLM L6 v2** | | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-All_MiniLM_L6_v2.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n\n### 语音转文本（STT）笔记本\n| 模型 | 类型 | 笔记本链接 |\n| --- | --- | --- |\n| **Whisper** **(Large)** | 微调 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Whisper.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n\n### OCR笔记本\n| 模型 | 类型 | 笔记本链接 |\n| --- | --- | --- |\n| **Deepseek OCR** **(3B)** | 微调 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Deepseek_OCR_(3B).ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **Deepseek OCR** **(3B)** | 评估 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Deepseek_OCR_(3B)-Evaluation.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **Deepseek OCR** **(3B)** | 评测 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Deepseek_OCR_(3B)-Eval.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **Deepseek OCR 2** **(3B)** | | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Deepseek_OCR_2_(3B).ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **Paddle OCR** **(1B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Paddle_OCR_(1B)_Vision.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n\n### BERT笔记本\n| 模型 | 类型 | 笔记本链接 |\n| --- | --- | --- |\n| **ModernBert** | | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-ModernBert.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **ModernBERT** **(Large)** | 分类 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-bert_classification.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n\n### Deepseek笔记本\n| 模型 | 类型 | 笔记本链接 |\n| --- | --- | --- |\n| **Deepseek OCR** **(3B)** | 微调 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Deepseek_OCR_(3B).ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **Deepseek OCR** **(3B)** | 评估 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Deepseek_OCR_(3B)-Evaluation.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **Deepseek OCR** **(3B)** | 评测 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Deepseek_OCR_(3B)-Eval.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **Deepseek OCR 2** **(3B)** | | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Deepseek_OCR_2_(3B).ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n\n### ERNIE 笔记本\n| 模型 | 类型 | 笔记本链接 |\n| --- | --- | --- |\n| **ERNIE 4.5 VL 28B A3B PT** | 视觉 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-ERNIE_4_5_VL_28B_A3B_PT_Vision.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n| **ERNIE 4.5 21B A3B PT** | 对话 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-ERNIE_4_5_21B_A3B_PT-Conversational.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n\n### GPT-OSS 笔记本\n| 模型 | 类型 | 笔记本链接 |\n| --- | --- | --- |\n| **gpt oss MXFP4** **(20B)** | 推理 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-GPT_OSS_MXFP4_(20B)-Inference.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https6:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n| **gpt oss** **(20B)** | 微调 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-gpt_oss_(20B)_500K_Context_Fine_tuning.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n| **gpt oss** **(20B)** | 微调 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-gpt-oss-(20B)-Fine-tuning.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n| **gpt oss BNB** **(20B)** | 推理 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-GPT_OSS_BNB_(20B)-Inference.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n| **(A100) gpt oss** **(120B)** | 微调 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-gpt-oss-(120B)_A100-Fine-tuning.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n\n### Gemma 笔记本\n| 模型 | 类型 | 笔记本链接 |\n| --- | --- | --- |\n| **Gemma3** **(4B)** | 对话 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Gemma3_(4B).ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n| **(A100) Gemma3** **(27B)** | 对话 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Gemma3_(27B)_A100-Conversational.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n| **Gemma3** **(270M)** | 对话 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Gemma3_(270M).ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n| **Gemma3N** **(4B)** | 多模态 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Gemma3N_(4B)-Conversational.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n| **Gemma3N** **(4B)** | 音频 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Gemma3N_(4B)-Audio.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n| **Gemma3N** **(2B)** | 推理 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Gemma3N_(2B)-Inference.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n| **Gemma3** **(4B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Gemma3_(4B)-Vision.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n| **Gemma3N** **(4B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Gemma3N_(4B)-Vision.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n| **EmbeddingGemma** **(300M)** | | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-EmbeddingGemma_(300M).ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n| **Gemma2** **(9B)** | Alpaca | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Gemma2_(9B)-Alpaca.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n| **Gemma2** **(2B)** | Alpaca | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Gemma2_(2B)-Alpaca.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n| **CodeGemma** **(7B)** | 对话 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-CodeGemma_(7B)-Conversational.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n\n### 石英笔记本\n| 模型 | 类型 | 笔记本链接 |\n| --- | --- | --- |\n| **石英4.0** **(3B)** | 对话型 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Granite4.0.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **石英4.0** **(350M)** | 对话型 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Granite4.0_350M.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n\n### 混合注意力笔记本\n| 模型 | 类型 | 笔记本链接 |\n| --- | --- | --- |\n| **液态LFM2** **(1.2B)** | 对话型 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Liquid_LFM2_(1.2B)-Conversational.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **猎鹰H1** **(0.5B)** | 阿尔帕卡 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Falcon_H1_(0.5B)-Alpaca.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n\n### 羊驼笔记本\n| 模型 | 类型 | 笔记本链接 |\n| --- | --- | --- |\n| **羊驼3.1** **(8B)** | 推理 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Llama3.1_(8B)-Inference.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **羊驼3** **(8B)** | Ollama | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Llama3_(8B)-Ollama.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **羊驼3** **(8B)** | 阿尔帕卡 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Llama3_(8B)-Alpaca.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **羊驼3.2** **(1B和3B)** | 对话型 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Llama3.2_(1B_and_3B)-Conversational.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **羊驼3.2** **(1B)** | RAFT | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Llama3.2_(1B)-RAFT.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **羊驼3.1** **(8B)** | 阿尔帕卡 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Llama3.1_(8B)-Alpaca.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **(A100) 羊驼3.3** **(70B)** | 对话型 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Llama3.3_(70B)_A100-Conversational.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **羊驼3.2** **(11B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Llama3.2_(11B)-Vision.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **羊驼3** **(8B)** | 对话型 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Llama3_(8B)-Conversational.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **小羊驼** **(1.1B)** | 阿尔帕卡 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-TinyLlama_(1.1B)-Alpaca.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n\n### Mistral 笔记本\n| 模型 | 类型 | 笔记本链接 |\n| --- | --- | --- |\n| **Mistral v0.3** **(7B)** | 对话型 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Mistral_v0.3_(7B)-Conversational.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n| **Magistral** **(24B)** | 推理对话型 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Magistral_(24B)-Reasoning-Conversational.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n| **Pixtral** **(12B)** | 视觉型 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Pixtral_(12B)-Vision.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n| **Mistral Small** **(22B)** | Alpaca 型 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Mistral_Small_(22B)-Alpaca.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n| **Ministral3 VL** **(3B)** | 视觉型 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Ministral_3_VL_(3B)_Vision.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n| **Mistral v0.3** **(7B)** | Alpaca 型 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Mistral_v0.3_(7B)-Alpaca.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n| **Mistral Nemo** **(12B)** | Alpaca 型 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Mistral_Nemo_(12B)-Alpaca.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n\n### Nemotron 笔记本\n| 模型 | 类型 | 笔记本链接 |\n| --- | --- | --- |\n| **(A100) Nemotron Nano 3 30B A3B** | | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Nemotron-Nano-3-30B-A3B_A100.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n| **(A100) Nemotron 3 Nano 30B A3B** | | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Nemotron-3-Nano-30B-A3B_A100.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n\n### Paddle 笔记本\n| 模型 | 类型 | 笔记本链接 |\n| --- | --- | --- |\n| **Paddle OCR** **(1B)** | 视觉型 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Paddle_OCR_(1B)_Vision.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n\n### Phi 笔记本\n| 模型 | 类型 | 笔记本链接 |\n| --- | --- | --- |\n| **Phi 4** | 对话型 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Phi_4-Conversational.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n| **Phi 3.5 Mini** | 对话型 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Phi_3.5_Mini-Conversational.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n| **Phi 3 Medium** | 对话型 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Phi_3_Medium-Conversational.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在 Kaggle 中打开\">\u003C\u002Fa> |\n\n### 通义千问笔记本\n| 模型 | 类型 | 笔记本链接 |\n| --- | --- | --- |\n| **Qwen3** **(4B)** | QAT | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen3_(4B)_Instruct-QAT.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https6:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **Qwen3** **(4B)** | 对话 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen3_(4B)-Instruct.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **Qwen2.5 VL** **(7B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen2.5_VL_(7B)-Vision.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **Qwen3 VL** **(8B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen3_VL_(8B)-Vision.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **(A100) Qwen3** **(32B)** | 理性对话 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen3_(32B)_A100-Reasoning-Conversational.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **Qwen3** **(14B)** | 理性对话 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen3_(14B)-Reasoning-Conversational.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **Qwen3** **(14B)** | | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen3_(14B).ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **Qwen2 VL** **(7B)** | 视觉 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen2_VL_(7B)-Vision.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **Qwen3** **(4B)** | 思考 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen3_(4B)-Thinking.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **Qwen2.5** **(7B)** | 阿尔帕卡 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen2.5_(7B)-Alpaca.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **Qwen2.5 Coder** **(14B)** | 对话 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen2.5_Coder_(14B)-Conversational.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **Qwen3 Embedding** **(4B)** | | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen3_Embedding_(4B).ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **Qwen3 Embedding** **(0.6B)** | | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen3_Embedding_(0.6B).ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **Qwen3** **(14B)** | 阿尔帕卡 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen3_(14B)-Alpaca.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **Qwen2** **(7B)** | 阿尔帕卡 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Qwen2_(7B)-Alpaca.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n\n### 文本补全 \u002F 继续预训练笔记本\n| 模型 | 类型 | 笔记本链接 |\n| --- | --- | --- |\n| **Mistral v0.3** **(7B)** | CPT | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Mistral_v0.3_(7B)-CPT.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **Mistral** **(7B)** | 文本补全 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Mistral_(7B)-Text_Completion.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n\n### 其他笔记本\n| 模型 | 类型 | 笔记本链接 |\n| --- | --- | --- |\n| **CodeForces cot 微调\u003Cbr>用于 CodeForces 上的推理** | 推理 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-CodeForces-cot-Finetune_for_Reasoning_on_CodeForces.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n| **Unsloth** | 工作室 | \u003Ca href=\"https:\u002F\u002Fwww.kaggle.com\u002Fnotebooks\u002Fwelcome?src=https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnb\u002FKaggle-Unsloth_Studio.ipynb&accelerator=nvidiaTeslaT4\" target=\"_blank\" rel=\"noopener noreferrer\">\u003Cimg src=\"https:\u002F\u002Fkaggle.com\u002Fstatic\u002Fimages\u002Fopen-in-kaggle.svg\" alt=\"在Kaggle中打开\">\u003C\u002Fa> |\n\n\u003C\u002Fdetails>\n\n\n\u003C!-- 笔记本链接结束 -->\n\n## 已知问题 \u002F 环境说明\n\n- **NumPy 2.x ↔ soxr**: NumPy 2.x 会破坏 soxr，导致 Unsloth 导入失败。可通过固定 `numpy\u003C2` 来解决。如果需要，可以使用 `pip install --force-reinstall \"numpy\u003C2\"`。_影响：无法运行 Unsloth。_\n- **soxr 重新安装**：如果不使用 `--no-deps`，`pip install --force-reinstall soxr` 可能会再次引入 NumPy 2.x。应使用 `pip install --force-reinstall --no-deps soxr` 来避免此问题。_影响：可能重新引入 NumPy 2.x 并破坏 Unsloth 的导入。_\n- **typing_extensions**：较旧版本的 typing_extensions 可能会导致 torch 导入失败（缺少 TypeIs），直到升级为止。请使用 `pip install --upgrade typing_extensions` 进行升级。_影响：无法正确导入 PyTorch。_\n- **解析器警告**：固定 `numpy\u003C2` 可能会在与 SciPy\u002FNumba 一起使用时引发 pip 解析器警告；通常这些警告不会造成严重问题。_影响：仅为外观上的警告，不影响功能。_\n- **ROCm \u002F triton_key**：在 ROCm 上使用 `torch.compile` 时，如果 Triton 缺少 `triton_key`，LoRA 的反向传播可能会崩溃。目前的解决方法是在 ROCm 上禁用 Inductor\u002Fcompile（代码中已处理，但仍需注意）。_影响：在使用 torch.compile 时，可能导致 AMD GPU 上的训练崩溃。_\n\n# ✨ 参与笔记本贡献\n\n如果您想为我们的笔记本贡献内容，以下是一个入门指南：\n\n1. **找到模板**：我们在该项目的根目录下提供了一个名为 `Template_Notebook.ipynb` 的模板笔记本。该模板包含了本系列所有笔记本的基本结构和格式规范。\n2. **创建您的笔记本**：\n * 复制 `Template_Notebook.ipynb`。\n * 按照以下命名规则重命名复制后的文件：\n * **LLM 笔记本**：`\u003C模型名称>-\u003C类型>.ipynb`（例如：`Mistral_v0.3_(7B)-Alpaca.ipynb`）\n * **视觉笔记本**：`\u003C模型名称>-Vision.ipynb`（例如：`Llava_v1.6_(7B)-Vision.ipynb`）\n * **类型示例**：`Alpaca`、`对话式`、`CPT`、`DPO`、`ORPO`、`文本补全`、`CSV`、`推理`、`Unsloth_Studio`\n \u003C!-- * 修改您笔记本的内容，添加代码、解释以及其他相关信息。请务必遵循模板中的结构和指南。 -->\n3. **放入 `original_template` 目录**：当您的笔记本准备就绪后，将其移动到 `original_template` 目录。\n4. **更新笔记本**：在终端中运行以下命令：\n ```bash\n python update_all_notebooks.py\n ```\n 该脚本将自动执行以下操作：\n * 将您的笔记本从 `original_template` 目录复制到 `notebooks` 目录。\n * 更新笔记本内部的各个部分（如安装、新闻等），以确保一致性。\n * 将您的笔记本添加到本 `README.md` 文件中的相应列表中。\n5. **创建拉取请求**：完成后，只需创建一个拉取请求（PR）以合并您的更改，使您的内容对所有人可见！\n * 我们非常感谢您的贡献，并期待审阅您的笔记本！","# Unsloth Notebooks 快速上手指南\n\nUnsloth Notebooks 是 Unsloth 团队提供的一系列开源 Jupyter Notebook，旨在帮助开发者在 Google Colab 或本地环境中快速对主流大语言模型（如 Llama 3、Qwen、Gemma 等）进行微调、推理及强化学习训练。本指南将协助您快速开始使用。\n\n## 环境准备\n\n### 系统要求\n* **运行平台**：推荐使用 **Google Colab**（免费或 Pro 版本），也可在本地 Linux 环境运行。\n* **GPU 要求**：\n * 微调小型模型（1B-4B）：建议至少 16GB VRAM（如 T4, A10G）。\n * 微调中型模型（7B-14B）：建议 24GB+ VRAM（如 A100, H100, RTX 3090\u002F4090）。\n * 部分强化学习（GRPO）示例可能需要更高显存。\n* **本地环境依赖**：\n * Python 3.10+\n * CUDA 12.1+ (本地运行时)\n * Git\n\n### 前置依赖\n若选择在本地运行，需确保已安装基础深度学习库。Unsloth 会自动处理大部分依赖，但建议预先检查 NVIDIA 驱动状态。\n\n## 安装步骤\n\n### 方案一：直接使用 Google Colab（推荐）\n无需手动安装，直接点击 README 中的 **\"Open in Colab\"** 按钮即可。Notebook 内部包含自动安装脚本。\n\n### 方案二：本地环境安装\n如果您希望在本地复现 Notebook 流程，请执行以下命令安装 Unsloth 核心库。\n\n**注意**：国内用户建议使用清华源或阿里源加速 pip 下载。\n\n```bash\n# 1. 升级 pip\npip install --upgrade pip\n\n# 2. 安装 Unsloth (包含最新优化内核)\n# 国内加速示例 (使用清华源)\npip install \"unsloth[colab-new] @ git+https:\u002F\u002Fgithub.com\u002Funslothai\u002Funsloth.git\" \\\n -i https:\u002F\u002Fpypi.tuna.tsinghua.edu.cn\u002Fsimple\n\n# 3. 安装额外依赖 (根据具体模型需求，通常在 Notebook 中会自动执行)\npip install xformers trl peft accelerate bitsandbytes datasets -i https:\u002F\u002Fpypi.tuna.tsinghua.edu.cn\u002Fsimple\n```\n\n> **提示**：本地运行时，请确保您的 CUDA 版本与 PyTorch 版本兼容。Unsloth 对特定 CUDA 版本有优化，如遇报错请参考官方文档切换 CUDA 版本。\n\n## 基本使用\n\n最简单的使用方式是直接运行官方提供的预置 Notebook。以下是基于 **Llama 3.1 (8B)** 进行微调的通用流程示例（逻辑适用于所有列表中的模型）。\n\n### 步骤 1: 加载模型与分词器\n在 Notebook 单元格中运行以下代码初始化模型。Unsloth 会自动加载量化版本以节省显存。\n\n```python\nfrom unsloth import FastLanguageModel\n\nmax_seq_length = 2048 # 支持长上下文，RoPE 缩放\ndtype = None # 自动检测 (bf16 或 fp16)\nload_in_4bit = True # 启用 4-bit 量化以节省显存\n\nmodel, tokenizer = FastLanguageModel.from_pretrained(\n model_name = \"unsloth\u002FLlama-3.1-8B\", # 可替换为 Qwen, Gemma 等\n max_seq_length = max_seq_length,\n dtype = dtype,\n load_in_4bit = load_in_4bit,\n)\n```\n\n### 步骤 2: 配置 LoRA 适配器\n设置低秩适配器（LoRA）参数，仅微调少量参数，大幅降低显存占用。\n\n```python\nmodel = FastLanguageModel.get_peft_model(\n model,\n r = 16, # LoRA 秩\n target_modules = [\"q_proj\", \"k_proj\", \"v_proj\", \"o_proj\",\n \"gate_proj\", \"up_proj\", \"down_proj\"],\n lora_alpha = 16,\n lora_dropout = 0,\n bias = \"none\",\n use_gradient_checkpointing = \"unsloth\", # 节省显存的梯度检查点\n random_state = 3407,\n use_rslora = False, \n loftq_config = None,\n)\n```\n\n### 步骤 3: 准备数据并训练\n定义数据集格式并启动训练。\n\n```python\nfrom trl import SFTTrainer\nfrom transformers import TrainingArguments\nfrom unsloth import is_bfloat16_supported\n\ntrainer = SFTTrainer(\n model = model,\n tokenizer = tokenizer,\n train_dataset = your_dataset, # 替换为您的数据集\n dataset_text_field = \"text\",\n max_seq_length = max_seq_length,\n dataset_num_proc = 2,\n packing = False, \n args = TrainingArguments(\n per_device_train_batch_size = 2,\n gradient_accumulation_steps = 4,\n warmup_steps = 5,\n max_steps = 60,\n learning_rate = 2e-4,\n fp16 = not is_bfloat16_supported(),\n bf16 = is_bfloat16_supported(),\n logging_steps = 1,\n output_dir = \"outputs\",\n optim = \"adamw_8bit\",\n seed = 3407,\n ),\n)\n\ntrainer_stats = trainer.train()\n```\n\n### 步骤 4: 推理测试\n训练完成后，立即测试模型效果。\n\n```python\nFastLanguageModel.for_inference(model)\ninputs = tokenizer([\"你好，请介绍一下你自己。\"], return_tensors = \"pt\").to(\"cuda\")\noutputs = model.generate(**inputs, max_new_tokens = 64, use_cache = True)\nprint(tokenizer.batch_decode(outputs))\n```\n\n对于视觉模型（Vision）、音频模型或强化学习（GRPO）任务，请直接访问 README 表格中对应的 **Notebook Link**，其中已包含针对该模态的完整数据处理和训练代码。","一家初创教育科技公司希望快速构建一个能理解数学公式并讲解解题步骤的专属多模态 AI 助教，团队仅有一名算法工程师且算力资源有限。\n\n### 没有 notebooks 时\n- **环境配置耗时极长**：工程师需手动排查 PyTorch、CUDA 版本兼容性，安装视觉与语言模型的复杂依赖库往往耗费数天时间。\n- **代码复用率低**：从数据预处理、QLoRA 微调到推理验证，每个环节都需从零编写样板代码，极易出现难以调试的底层错误。\n- **多模态适配困难**：让模型同时处理图像（题目截图）和文本（解题逻辑）需要深厚的架构知识，自行修改注意力机制风险极高。\n- **试错成本高昂**：在本地或云端反复尝试不同参数组合时，因缺乏优化导致显存溢出，频繁中断训练进程。\n\n### 使用 notebooks 后\n- **一键启动开发**：直接打开针对 Qwen3.5-VL 等视觉模型的 Colab 笔记本，预置好所有依赖与环境，几分钟内即可开始数据加载。\n- **全流程模板化**：利用内置的数据清洗、Unsloth 加速训练及推理代码块，工程师只需替换数据集路径即可跑通完整微调链路。\n- **多模态能力开箱即用**：调用现成的 Vision 类型笔记本，无需修改底层架构即可实现“看图说话”的数学题讲解功能。\n- **高效资源利用**：借助笔记中集成的显存优化技术，在单张 T4 GPU 上也能流畅训练 8B 甚至更大参数的多模态模型。\n\nnotebooks 将原本需要数周的环境搭建与原型验证工作压缩至几小时，让小型团队也能低成本落地垂直领域的多模态大模型应用。","https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Funslothai_notebooks_9c07eafd.png","unslothai","Unsloth AI","https:\u002F\u002Foss.gittoolsai.com\u002Favatars\u002Funslothai_6d372387.png","Run and train AI models locally. 🦥",null,"https:\u002F\u002Funsloth.ai\u002F","https:\u002F\u002Fgithub.com\u002Funslothai",[80,84,88],{"name":81,"color":82,"percentage":83},"Jupyter Notebook","#DA5B0B",95.7,{"name":85,"color":86,"percentage":87},"Python","#3572A5",4.3,{"name":89,"color":90,"percentage":91},"PowerShell","#012456",0,5237,848,"2026-04-17T07:03:24","LGPL-3.0","Linux, macOS, Windows","需要 NVIDIA GPU（推荐 T4, A100, H100 或本地 RTX 3090\u002F4090），显存需求视模型大小而定（2B 模型约 6GB+，20B 模型需 24GB+ 或使用量化\u002FOffload 技术），支持 CUDA 11.8+ 及 BF16\u002FFP8 精度","最低 12GB，推荐 24GB+（取决于加载的模型参数量）",{"notes":100,"python":101,"dependencies":102},"该仓库主要提供 Google Colab 笔记本，可直接在云端免费运行（需绑定 GPU 运行时）。若在本地运行，建议安装 'unsloth' 库以启用针对 Llama、Mistral、Gemma 等模型的 2 倍加速和内存优化。部分高级功能（如 GRPO 强化学习、FP8 训练）需要较新的 GPU 架构（如 Ampere 或 Hopper 系列）。","3.9+",[103,104,105,106,107,108,109,110],"unsloth","torch","transformers","accelerate","xformers","bitsandbytes","trl","peft",[35,15,112,14,113],"音频","其他",[103],"2026-03-27T02:49:30.150509","2026-04-18T00:47:15.732972",[118,123,128,133,138,143,148],{"id":119,"question_zh":120,"answer_zh":121,"source_url":122},38245,"在免费层级运行 Gemma3 (270M) Notebook 时遇到 'OutOfResources: out of resource: shared memory' 错误怎么办？","该错误通常是因为免费层级的共享内存限制（65536）低于模型需求。虽然尝试减少 batch size 或使用 4bit 加载可能无效，但建议检查是否使用了正确的运行时环境。如果问题持续，可能需要升级到更高配置的实例或等待官方对 Notebook 进行优化以适配免费层级资源限制。","https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fissues\u002F87",{"id":124,"question_zh":125,"answer_zh":126,"source_url":127},38246,"如何在 Unsloth 中微调 OuteTTS 或其他非标准架构模型（如 CSM）？","Unsloth 社区已通过 Pull Request #34 添加了对 OuteTTS 模型的支持。对于其他简单架构模型（如 CSM），如果其训练代码结构简单，理论上可以尝试使用 Unsloth 进行加速，但需确认模型架构是否被底层库完全支持。建议查看已合并的 PR 代码作为参考来实现自定义模型的微调。","https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fissues\u002F32",{"id":129,"question_zh":130,"answer_zh":131,"source_url":132},38247,"使用 Llama 3.2 Notebook 微调自定义数据集时，为什么聊天模板（chat_template）显示错误或混乱？","这通常是因为选择了错误的基模型版本。请确保使用的是 'Instruct' 版本的模型（例如 3B-Instruct），而不是普通的基座模型（Base Model）。基座模型缺乏特定的聊天模板标记，会导致输出格式混乱。切换为 Instruct 模型即可解决此问题。","https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fissues\u002F140",{"id":134,"question_zh":135,"answer_zh":136,"source_url":137},38248,"GitHub 上的 Notebook 文件无法在浏览器中直接渲染或预览怎么办？","如果 Notebook 在 GitHub 页面上无法渲染，请直接下载该 .ipynb 文件，然后上传到 Google Colab 或 Kaggle 环境中运行。官方仓库的 README.md 通常也提供了直接跳转到 Colab 或 Kaggle 的链接，建议使用这些链接以获得最佳体验。","https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fissues\u002F31",{"id":139,"question_zh":140,"answer_zh":141,"source_url":142},38249,"在 Kaggle 上运行 Orpheus TTS 微调 Notebook 时，执行 trainer.train() 报错 ValueError 或数据加载错误如何解决？","该问题通常由 Notebook 版本过旧或依赖冲突引起。维护者已通过 Pull Request #27 修复了此问题。请确保你使用的是最新版本的 Notebook 代码，或者手动应用 PR #27 中的修改来解决数据迭代器和损失计算相关的错误。","https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fissues\u002F24",{"id":144,"question_zh":145,"answer_zh":146,"source_url":147},38250,"运行 Qwen2-VL (7B) 视觉模型 Notebook 时，输出大量重复的 'system' 文本且微调失败是什么原因？","这是 Unsloth 库早期版本中的一个已知 Bug，导致视觉模型在处理特定输入时产生重复输出。该问题已在 Unsloth 主仓库的 Issue #2120 中得到修复。请确保你安装了最新版本的 unsloth 库（pip install --upgrade unsloth），然后重新运行 Notebook。","https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fissues\u002F13",{"id":149,"question_zh":150,"answer_zh":151,"source_url":152},38251,"Qwen3 Reasoning Notebook 中的数据采样比例与文档描述不一致（实际推理数据占比过高）如何修正？","这是一个代码逻辑错误。文档意图是 25% 推理数据和 75% 聊天数据，但原代码错误地计算了采样数量。修正方法是将采样代码调整为：先确定推理数据的数量，然后根据目标比例计算所需的聊天数据数量。具体代码逻辑应确保 non_reasoning_subset 的采样数量为 reasoning_conversations 数量的 3 倍（即 75% vs 25%），而不是当前的反向逻辑。","https:\u002F\u002Fgithub.com\u002Funslothai\u002Fnotebooks\u002Fissues\u002F38",[]]