大语言模型

大语言模型(国外)分类当前收录 247 个开源模型与基建项目,按 S-Score 排名靠前的有 llama(S·84)、FinGPT(S·84)、grok-1(S·84)。S-Score 基于 GitHub 星标量级与维护活跃度算法计算,每小时重排、付费不影响排名,数据每日快照。本分类属于「模型 · 基建」板块,不进入 Skill 主榜,总览见 模型看板。

#SKILL · 摊位S-SCORESTARS ★ 7D趋势

1llamaMeta Llama 2,开源大模型里程碑S·84—★ 59.6k+20☆

2FinGPT开源金融领域大语言模型S·84—★ 21.3k+29☆

3grok-1xAI开源Grok-1，314B参数MoE大模型S·84—★ 52.2k☆

4mamba状态空间模型Mamba，线性复杂度挑战TransformerS·81—★ 18.8k+21☆

5llama3Meta Llama 3官方仓库,开源模型新标杆S·81—★ 29.2k☆

6MOSS复旦大学开源的工具增强对话大模型S·80—★ 12.2k+5☆

AD◧广告位招商中 —— 你的产品出现在这里自助投放 · 周结算 · 广告内容与榜单排序无关了解 →广告 AD

7HRM官方发布的分层推理模型S·79—★ 12.6k+6☆

8mistral-inferenceMistral AI 官方模型推理库,开源欧洲之光S·78—★ 10.8k☆

9ChatRWKV基于RWKV架构的开源对话模型，类ChatGPT体验S·78—★ 9.5k☆

10StableLMStability AI开源语言模型系列S·77—★ 15.7k☆

11gemmaGoogle Gemma 开源轻量模型,同源 Gemini 技术S·76—★ 5.7k+15☆

12GLiNER轻量级通用命名实体识别模型,支持任意实体类型S·75—★ 3.6k+30☆

13dollyDatabricks开源大语言模型S·75—★ 10.8k+4☆

14llama-modelsMeta Llama 全系列官方模型仓库S·74—★ 7.7k+1☆

15Yi01-ai团队从零训练的开源大模型系列S·74—★ 7.8k☆

16OLMoAllenAI 完全开放大模型,数据训练全公开S·73—★ 6.7k+8☆

17BitNet1比特Transformer大模型的PyTorch实现S·72—★ 1.9k+1☆

18openchat开源大语言模型，用不完美数据训练对话能力S·72—★ 5.5k☆

19picoGPT用NumPy实现的极简GPT-2教学项目S·70—★ 3.5k+2☆

20KnowLM开源知识增强大语言模型框架S·65—★ 1.4k☆

21xLAM面向AI Agent的大型动作模型系列S·64—★ 638☆

22Aquila2智源研究院发布的Aquila2大模型系列S·59—★ 445☆

23generative-ai-for-beginners微软官方 21 课生成式 AI 入门,零基础起步S·93—★ 119.7k+375☆

24LLMs-from-scratch用 PyTorch 从零手写 ChatGPT 级 LLM,逐步实现S·92—★ 105.0k+435☆

25ai-engineering-from-scratch从零学AI工程,学会做出来交付出去S·88▼1★ 54.6k+1764☆

26DeepTutor终身个性化AI辅导平台S·87—★ 39.6k+636☆

27awesome-generative-ai-guide生成式AI一站式学习资源大全S·86—★ 29.4k+130☆

28free-llm-api-resources免费LLM API资源汇总清单S·86—★ 29.3k+667☆

29llm-courseLLM 入门课程:路线图 + Colab 笔记本S·86—★ 82.8k+413☆

30VaneAI驱动的智能问答引擎S·86—★ 36.8k+122☆

31book-to-skillTurn any technical book PDF into a Claude Code skill — ready to study, reference, and use while you work.S·86—★ 30.7k+1678☆

32Anthropic-Cybersecurity-Skills817 条网络安全技能库,映射六大安全框架S·85—★ 32.8k+409☆

33TencentDB-Agent-MemoryTencentDB Agent Memory delivers fully local long-term memory for AI Agents via a 4-tier progressive pipeline, with zero external API dependeS·85—★ 26.7k+575☆

34heretic全自动移除语言模型的审查限制S·85—★ 31.4k+568☆

35machine-learning-for-trading《机器学习交易》第三版配套代码库S·84—★ 20.9k+80☆

36generative-ai谷歌云生成式 AI 示例代码与教程合集S·83—★ 17.7k+29☆

37browser-harnessBrowser Harness | Self-healing harness that enables LLMs to complete any task.S·83—★ 17.5k+125☆

38ml-engineering机器学习工程免费开源电子书S·83—★ 19.0k+62☆

39opencodexUniversal provider proxy for OpenAI Codex & Claude Code — use any LLM (Claude, Gemini, Grok, DeepSeek, Ollama…) with Codex CLI, App, SDK, anS·82—★ 14.6k+817☆

40learn-harness-engineeringHarness engineering beginner tutorial, from 0 to 1S·81—★ 15.2k+281☆

41hermes-studioWeb dashboard for Hermes Agent — multi-platform AI chat, session management, scheduled jobs, usage analytics S·81—★ 11.1k+160☆

42llm-engineer-toolkit精选120余个大模型开发库分类清单S·80—★ 10.8k+1☆

43omnigentOmnigent is an open-source AI agent framework and meta-harness: orchestrate Claude Code, Codex, Cursor, Pi, and custom agents — swap harnessS·80—★ 10.0k+186☆

44GraftTurbocharge Claude Code, Cursor, Codex, Gemini & every coding agent: faster, cheaper, with contextual understanding specific to your codebasS·79—★ 7.8k+1788☆

45kimi-k3-in-cA 2.78-trillion-parameter Kimi K3 running inference on a single CPU in 8.24 GB of RAM. Portable C99: no BLAS, no framework, no GPU.S·79—★ 7.9k+740☆

46open-multi-agentTypeScript AI agent orchestration framework with dynamic workflows. Describe the goal, not the graph: a coordinator plans the task DAG at ruS·78—★ 6.9k+34☆

47apfelThe free AI already on your Mac. CLI tool, OpenAI-compatible server, and interactive chat — all on-device via Apple Intelligence. No API keyS·78—★ 6.4k+28☆

48turbo-fieldfareGemma 4 26B-A4B inference in ~2 GB of RAM on any M-series MacBookS·78—★ 6.7k+65☆

49KunLocal-first AI agent workspace for coding, writing, design, research, and automation — one runtime for desktop GUI and TUI.S·78—★ 6.3k+17☆

50opensquillaOpenSquilla — Token-Efficient AI Agent with same budget, higher intelligence densityS·78—★ 7.0k+41☆

51LLMsPracticalGuideLLM实践资源精选清单，涵盖知识树与论文S·78—★ 10.2k☆

52awesome-free-llm-apisList of Permanent Free LLM API (API Keys)S·78—★ 7.6k+193☆

53makaApache Maka (Incubating) is a local-first AI agent workspace. Model messages, tool calls, tool results, permission decisions, and terminatioS·77—★ 5.4k+468☆

54flash-linear-attention高效线性注意力机制实现库S·77—★ 5.7k+36☆

55adhdADHD — a skill for coding agents. Tree-of-thought with pruning, built on the Claude & Codex Agent SDK. Fans out parallel divergent thoughts S·76—★ 4.2k+92☆

56awesome-opensource-aiCurated list of the best truly open-source AI projects, models, tools, and infrastructure. Daily updated.S·76—★ 4.7k+34☆

57open-scienceOpen Science is an open-source, local-first, model-agnostic AI research workbench for scientific discovery.S·76—★ 4.2k+85☆

58reasoning-from-scratch用PyTorch从零实现推理型LLM，分步教学S·76—★ 5.2k+53☆

59GenerativeAIExamples面向加速硬件的生成式AI参考工作流集合S·76—★ 4.2k+5☆

60im-not-aiAI가 쓴 한글을 사람 글처럼 윤문하는 Claude 스킬 — Korean AI-text humanizer: detects and rewrites translationese, mechanical parallelism, and 71 other AI telS·76—★ 5.6k+242☆

展示前 60 名 · 完整榜单见 每日热榜 →