{"data":{"slug":"aihubcn-awesome-chinese-llm","name":"Awesome-Chinese-LLM","tagline":"整理开源的中文大语言模型","github_url":"https://github.com/AiHubCN/Awesome-Chinese-LLM","owner":"AiHubCN","repo":"Awesome-Chinese-LLM","owner_avatar_url":"https://avatars.githubusercontent.com/u/29895268?v=4","primary_language":null,"stars":22738,"forks":2134,"topics":["awesome-lists","chatglm","chinese","llama","llm","nlp"],"archived":false,"github_pushed_at":"2026-05-10T05:03:06+00:00","maintenance_label":"Slowing","stars_delta_30d":53,"url":"https://www.graphcanon.com/tools/aihubcn-awesome-chinese-llm","markdown_url":"https://www.graphcanon.com/tools/aihubcn-awesome-chinese-llm.md","api_url":"https://www.graphcanon.com/api/graphcanon/tools/aihubcn-awesome-chinese-llm","graph_url":"https://www.graphcanon.com/api/graphcanon/graph?tool=aihubcn-awesome-chinese-llm","description":"整理开源的中文大语言模型，以规模较小、可私有化部署、训练成本较低的模型为主，包括底座模型，垂直领域微调及应用，数据集与教程等。","homepage_url":null,"license":null,"open_issues":27,"watchers":246,"ai_summary":"汇集了规模较小、适合私有化部署和低成本训练的中文大语言模型，包括基础模型、垂直领域的微调应用、相关数据集与教程。","readme_excerpt":"\n
\n

Awesome Chinese LLM

\n \n
\n\n

\n\"Awesome-Chinese-LLM\"\n

\n

\n An Awesome Collection for LLM in Chinese \n

\n

\n 收集和梳理中文LLM相关 \n

\n

\n \"GitHub\n \"GitHub\n \"GitHub\n

\n\n自ChatGPT为代表的大语言模型（Large Language Model, LLM）出现以后，由于其惊人的类通用人工智能（AGI）的能力，掀起了新一轮自然语言处理领域的研究和应用的浪潮。尤其是以ChatGLM、LLaMA等平民玩家都能跑起来的较小规模的LLM开源之后，业界涌现了非常多基于LLM的二次微调或应用的案例。本项目旨在收集和梳理中文LLM相关的开源模型、应用、数据集及教程等资料，目前收录的资源已达100+个！\n\n如果本项目能给您带来一点点帮助，麻烦点个⭐️吧～\n\n同时也欢迎大家贡献本项目未收录的开源模型、应用、数据集等。提供新的仓库信息请发起PR，并按照本项目的格式提供仓库链接、star数，简介等相关信息，感谢~\n\n\n\n常见底座模型细节概览：\n| 底座 | 包含模型 | 模型参数大小 | 训练token数 | 训练最大长度 | 是否可商用 |\n|----------|---------------------------|-----------------|-------------|------------|------- |\n| ChatGLM | ChatGLM/2/3/4 Base&Chat | 6B | 1T/1.4 | 2K/32K | 可商用 |\n| LLaMA | LLaMA/2/3 Base&Chat | 7B/8B/13B/33B/70B | 1T/2T | 2k/4k | 部分可商用 |\n| Baichuan | Baichuan/2 Base&Chat | 7B/13B | 1.2T/1.4T | 4k | 可商用 |\n| Qwen | Qwen/1.5/2/2.5 Base&Chat&VL | 7B/14B/32B/72B/110B | 2.2T/3T/18T | 8k/32k | 可商用 |\n| BLOOM | BLOOM | 1B/7B/176B-MT | 1.5T | 2k | 可商用 |\n| Aquila | Aquila/2 Base/Chat | 7B/34B | - | 2k | 可商用 |\n| InternLM | InternLM/2/2.5 Base/Chat/VL | 7B/20B | - | 200k | 可商用 |\n| Mixtral | Base&Chat | 8x7B | - | 32k | 可商用 |\n| Yi | Base&Chat | 6B/9B/34B | 3T | 200k | 可商用 |\n| DeepSeek | Base&Chat | 1.3B/7B/33B/67B | - | 4k | 可商用 |\n| XVERSE | Base&Chat | 7B/13B/65B/A4.2B| 2.6T/3.2T | 8k/16k/256k | 可商用 |\n| MiniMax | MiniMax-Text-01/VL-01 | 456B(45.9B activated) | - | 1M/4M | 可商用 |\n\n## 目录\n\n- [目录](#目录)\n - [1. 模型](#1-模型)\n - [1.1 文本LLM模型](#11-文本llm模型)\n - [1.2 多模态LLM模型](#12-多模态llm模型)\n - [2. 应用](#2-应用)\n - [2.1 垂直领域微调](#21-垂直领域微调)\n - [医疗](#医疗)\n - [法律](#法律)\n - [金融](#金融)\n - [教育](#教育)\n - [科技](#科技)\n - [电商](#电商)\n - [网络安全](#网络安全)\n - [农业](#农业)\n - [2.2 LangChain应用](#22-langchain应用)\n - [2.3 其他应用](#23-其他应用)\n - [3. 数据集](#3-数据集)\n - [预训练数据集](#预训练数据集)\n - [SFT数据集](#sft数据集)\n - [偏好数据集](#偏好数据集)\n - [4. LLM训练微调框架](#4-llm训练微调框架)\n - [5. LLM推理部署框架](#5-llm推理部署框架)\n - [6. LLM评测](#6-llm评测)\n - [7. LLM教程](#7-llm教程)\n - [LLM基础知识](#llm基础知识)\n - [提示工程教程](#提示工程教程)\n - [LLM应用教程](#llm应用教程)\n - [LLM实战教程](#llm实战教程)\n - [8. 相关仓库](#8-相关仓库)\n- [Star History](#star-history)\n\n### 1. 模型\n\n#### 1.1 文本LLM模型\n\n* ChatGLM：\n * 地址：https://github.com/THUDM/ChatGLM-6B\n \n * 简介：中文领域效果最好的开源底座模型之一，针对中文问答和对话进行了优化。经过约 1T 标识符的中英双语训练，辅以监督微调、反馈自助、人类反馈强化学习等技术的加持\n* ChatGLM2-6B\n * 地址：https://github.com/THUDM/ChatGLM2-6B\n \n * 简介：基于开源中英双语对话模型 ChatGLM-6B 的第二代版本，在保留了初代模型对话流畅、部署门槛较低等众多优秀特性的基础之上，引入了GLM 的混合目标函数，经过了 1.4T 中英标识符的预训练与人类偏好对齐训练；基座模型的上下文长度扩展到了 32K，并在对话阶段使用 8K 的上下文长度训练；基于 Multi-Query Attention 技术实现更高效的推理速度和更低的显存占用；允许商业使用。\n* ChatGLM3-6B\n * 地址：https://github.com/THUDM/ChatGLM3\n \n * 简介：ChatGLM3-6B 是 ChatGLM3 系列中的开","github_created_at":"2023-05-22T12:27:03+00:00","created_at":"2026-07-07T17:33:00.13864+00:00","updated_at":"2026-08-17T00:01:46.974722+00:00","categories":[{"slug":"llm-frameworks","name":"LLM Frameworks","url":"https://www.graphcanon.com/categories/llm-frameworks","markdown_url":"https://www.graphcanon.com/categories/llm-frameworks.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/llm-frameworks"},{"slug":"model-training","name":"Model Training","url":"https://www.graphcanon.com/categories/model-training","markdown_url":"https://www.graphcanon.com/categories/model-training.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/model-training"}],"tags":[{"slug":"awesome-lists","name":"awesome-lists"},{"slug":"chatglm","name":"chatglm"},{"slug":"chinese","name":"chinese"},{"slug":"llama","name":"llama"},{"slug":"llm","name":"llm"},{"slug":"nlp","name":"nlp"}],"trust":{"provenance":{"is_fork":false,"github_id":643916827,"owner_type":"User","methodology":"github_public_v1","parent_repo":null,"near_duplicate_slugs":[]},"computed_at":"2026-08-17T00:01:46.262Z","maintenance":{"label":"Slowing","score":36,"methodology":"github_public_v1","releases_90d":0,"days_since_push":98,"last_release_at":null,"stars_delta_30d":53,"open_issues_delta_30d":3},"security_summary":{"status":"no_lockfile","scanner":null,"low_count":0,"high_count":0,"last_scan_at":"2026-07-11T11:00:05.894Z","medium_count":0,"scan_profile":"none","critical_count":0}},"capability_facts":{"scan":{"source":"repo_scan","observed_at":"2026-08-17T00:01:46.705Z"}},"decision_facts":{"hosting":null,"pricing":null,"requirements":null,"constraints":null,"when_to_use":["If you are looking to implement low-cost and efficient Chinese NLP solutions that can be deployed privately.","When your project requires access to both foundational and specific domain-tuned models along with supporting data sets and tutorials."],"when_not_to_use":["Avoid if your project necessitates large-scale, highly advanced computational capabilities or you are working with languages other than Chinese.","If your deployment scenario is limited to public cloud services only without the option for private deployment."],"source":"enrich:decision_facts","observed_at":"2026-07-11T13:58:24.376Z"},"constraint_facets":null,"decision_summary":[{"label":"Adopt for","value":"Awesome-Chinese-LLM is a curated list focusing on smaller, less computationally expensive Chinese language models suitable for private deployment."}]}}