The foundation layer. Language models, training frameworks, inference servers, embedding pipelines, evaluation kits, fine-tuning toolchains, and the supporting infrastructure that makes those things deployable. The /best/open-source-llms page is the curated top-N; this is the wide cut, mixing model releases, library releases, and infrastructure releases on equal footing because teams adopt them as a stack. Frontier weight drops spike every signal source in the same 48 hours, then settle. Quieter projects — inference servers, evaluation suites, RAG toolkits — hold position more durably than the spike curve suggests. The score weights long-tail engagement so both shapes rank in their right relative position. When picking what to invest in, the things that matter: licence terms on weights versus source (the two often differ — a permissive repo licence doesn't mean permissive weights), evaluation methodology (open leaderboards lie about what you'll see in production), and the maintainer's track record on minor versions.
| # | Repository | Stars | 24h | 7d | 30d | Trend | Mentions | Actions |
|---|---|---|---|---|---|---|---|---|
| 01 | NousResearch/hermes-agent The agent that grows with you | 191K | +6+0.0% | +7.9K+4.3% | +43K+29.0% |
| 02 | p-e-w/heretic Fully automatic censorship removal for language models | 24.2K | +91+0.4% | +422+1.8% | +3.4K+16.6% |
| 03 | github/spec-kit 💫 Toolkit to help you get started with Spec-Driven Development | 111.5K | +3+0.0% | +2.3K+2.1% | +13.4K+13.7% |
| 04 | Fincept-Corporation/FinceptTerminal FinceptTerminal is a modern finance application offering advanced market analytics, investment research, and economic data tools, designed for interactive exploration and data-driven decision-making in a user-friendly environment. | 26.4K | +22+0.1% | +967+3.8% | +5.5K+26.3% |
| 05 | openclaw/openclaw Your own personal AI assistant. Any OS. Any Platform. The lobster way. 🦞 | 378.2K | +1+0.0% | +1.1K+0.3% | +8.4K+2.3% |
| 06 | jlcodes99/cockpit-tools 🚀 通用 AI IDE 账号管理工具:支持 Antigravity / Codex / GitHub Copilot / Windsurf / Kiro / Cursor / Gemini-cli / CodeBuddy,多账号切换、配额监控、自动唤醒与多开实例管理。 🚀 Universal AI IDE account manager for Antigravity / Codex / GitHub Copilot / Windsurf / Kiro / Cursor / Gemini-cli / CodeBuddy, with multi-account switching, quota monitoring, wake-up automation, and multi-insta | 11.2K | +1+0.0% | +908+8.9% | +3.6K+48.2% |
| 07 | jundot/omlx LLM inference server with continuous batching & SSD caching for Apple Silicon — managed from the macOS menu bar | 16.4K | +2+0.0% | +473+3.0% | +2.6K+19.0% |
| 08 | maximhq/bifrost Fastest enterprise AI gateway (50x faster than LiteLLM) with adaptive load balancer, cluster mode, guardrails, 1000+ models support & <100 µs overhead at 5k RPS. | 5.7K | +3+0.1% | +174+3.2% | +804+16.5% |
| 09 | charmbracelet/crush Glamourous agentic coding for all 💘 | 25.2K | +2+0.0% | +196+0.8% | +1.1K+4.5% |
| 10 | firecrawl/firecrawl The API to search, scrape, and interact with the web at scale. 🔥 | 131.6K | +1+0.0% | +2.4K+1.9% | +12.4K+10.4% |
| 11 | BerriAI/litellm Python SDK, Proxy Server (AI Gateway) to call 100+ LLM APIs in OpenAI (or native) format, with cost tracking, guardrails, loadbalancing and logging. [Bedrock, Azure, OpenAI, VertexAI, Cohere, Anthropic, Sagemaker, HuggingFace, VLLM, NVIDIA NIM] | 50.1K | -7-0.0% | +665+1.3% | +3.3K+7.1% |
| 12 | qdrant/qdrant Qdrant - High-performance, massive-scale Vector Database and Vector Search Engine for the next generation of AI. Also available in the cloud https://cloud.qdrant.io/ | 32K | +1+0.0% | +215+0.7% | +820+2.6% |
| 13 | iOfficeAI/AionUi Free, local, open-source 24/7 Cowork app for OpenClaw, Hermes Agent, Claude Code, Codex, OpenCode, Gemini CLI and 20+ more CLI | Customize your assistants | Star if you like it! | 28.1K | -193-0.7% | +432+1.6% | +3.2K+12.8% |
| 14 | run-llama/liteparse A fast, helpful, and open-source document parser | 9.9K | +3+0.0% | +678+7.4% | +4.8K+94.5% |
| 15 | OpenBMB/VoxCPM VoxCPM2: Tokenizer-Free TTS for Multilingual Speech Generation, Creative Voice Design, and True-to-Life Cloning | 28.5K | +1+0.0% | +2K+7.7% | +9.9K+53.1% |
| 16 | Tencent/WeKnora Open-source LLM knowledge platform: turn raw documents into a queryable RAG, an autonomous reasoning agent, and a self-maintaining Wiki. | 16.2K | +1+0.0% | +180+1.1% | +1.4K+9.3% |
| 17 | shanraisshan/claude-code-best-practice from vibe coding to agentic engineering - practice makes claude perfect | 57.5K | +2+0.0% | +911+1.6% | +4.7K+9.0% |
| 18 | thedotmack/claude-mem Persistent Context Across Sessions for Every Agent – Captures everything your agent does during sessions, compresses it with AI, and injects relevant context back into future sessions. Works with Claude Code, OpenClaw, Codex, Gemini, Hermes, Copilot, OpenCode + More | 81.8K | +1+0.0% | +975+1.2% | +6.6K+8.8% |
| 19 | rohitg00/ai-engineering-from-scratch Learn it. Build it. Ship it for others. | 31.4K | +3+0.0% | +2.6K+9.0% | +23.9K+319.2% |
| 20 | openchamber/openchamber Desktop and web interface for OpenCode AI agent | 5.2K | — | +208+4.2% | +980+23.3% |
| 21 | junhoyeo/tokscale 🛰️ A CLI tool for tracking token usage from OpenCode, Claude Code, 🦞OpenClaw (Clawdbot/Moltbot), Pi, Codex, Gemini, Cursor, AmpCode, Factory Droid, Kimi, and more! • 🏅Global Leaderboard + 2D/3D Contributions Graph | 3.7K | +3+0.1% | +133+3.8% | +787+27.2% |
| 22 | danielmiessler/Fabric Fabric is an open-source framework for augmenting humans using AI. It provides a modular system for solving specific problems using a crowdsourced set of AI prompts that can be used anywhere. | 42.3K | +8+0.0% | +241+0.6% | +634+1.5% |
| 23 | NoFxAiOS/nofx Your AI trading terminal assistant for US stocks, commodities, forex, and crypto. | 12.5K | +1+0.0% | +46+0.4% | +159+1.3% |
| 24 | steipete/CodexBar Show usage stats for OpenAI Codex and Claude Code, without having to login. | 14.6K | +1+0.0% | +335+2.3% | +2.6K+21.1% |
| 25 | op7418/CodePilot A multi-model AI agent desktop client — connect any AI provider, extend with MCP & skills, control from your phone. Built with Electron + Next.js. | 6K | +2+0.0% | +75+1.3% | +315+5.5% |
| 26 | activeloopai/hivemind One brain for all your agents | 1.1K | — | +673+177.6% | +940+839.3% |
| 27 | apple/coremltools Core ML tools contain supporting tools for Core ML model conversion, editing, and validation. | 5.3K | +1+0.0% | +29+0.5% | +60+1.1% |
| 28 | blakeblackshear/frigate NVR with realtime local object detection for IP cameras | 33.7K | +1+0.0% | +140+0.4% | +1.8K+5.6% |
| 29 | opensquilla/opensquilla OpenSquilla — Token-Efficient AI Agent with same budget, higher intelligence density | 3.9K | +1+0.0% | +696+21.4% | +3.6K+1198.0% |
| 30 | wanshuiyin/Auto-claude-code-research-in-sleep ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent. | 12K | -65-0.5% | +478+4.2% | +2.8K+30.1% |
| 31 | alistaitsacle/free-llm-api-keys Free LLM API keys for GPT-5.5, Claude, DeepSeek, Gemini, Grok — copy, paste, use. Updated 3-5x daily. No credit card needed. | 2.2K | — | +667+42.7% | +2K+707.2% |
| 32 | OpenWhispr/openwhispr Voice-to-text dictation app with local (Nvidia Parakeet/Whisper) and cloud models (BYOK). Privacy-first and available cross-platform. | 3.7K | — | +123+3.4% | +631+20.5% |
| 33 | deepspeedai/DeepSpeed DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective. | 42.5K | — | +40+0.1% | +225+0.5% |
| 34 | cheahjs/free-llm-api-resources A list of free LLM inference resources accessible via API. | 23.3K | -23-0.1% | +380+1.7% | +1.9K+9.1% |
| 35 | linshenkx/prompt-optimizer An AI prompt optimizer for writing better prompts and getting better AI results. | 30.7K | +2+0.0% | +279+0.9% | +2.1K+7.3% |
| 36 | decolua/9router Unlimited FREE AI coding. Connect Claude Code, Codex, Cursor, Cline, Copilot, Antigravity to FREE Claude/GPT/Gemini via 40+ providers. Auto-fallback, RTK -40% tokens, never hit limits. | 17.3K | -950-5.2% | +807+4.9% | +7.5K+77.4% |
| 37 | cloudwego/eino The ultimate LLM/AI application development framework in Go. | 11.8K | -37-0.3% | +113+1.0% | +560+5.0% |
| 38 | cocoindex-io/cocoindex Incremental engine for long horizon agents 🌟 Star if you like it! | 10.3K | +1+0.0% | +73+0.7% | +595+6.2% |
| 39 | NVlabs/Sana SANA: Efficient High-Resolution Image Synthesis with Linear Diffusion Transformer | 8.2K | +1+0.0% | +80+1.0% | +3.1K+60.1% |
| 40 | deepset-ai/haystack Open-source AI orchestration framework for building context-engineered, production-ready LLM applications. Design modular pipelines and agent workflows with explicit control over retrieval, routing, memory, and generation. Built for scalable agents, RAG, multimodal applications, semantic search, and conversational systems. | 25.5K | — | +76+0.3% | +344+1.4% |
| 41 | alibaba/MNN MNN: A blazing-fast, lightweight inference engine battle-tested by Alibaba, powering high-performance on-device LLMs and Edge AI. | 15.5K | +1+0.0% | +67+0.4% | +332+2.2% |
| 42 | CodeWithCJ/SparkyFitness SparkyFitness: Built for Families. Powered by AI. Track food, fitness, water, and health — together. | 4.2K | — | +61+1.5% | +783+22.9% |
| 43 | OpenMOSS/MOSS-TTS MOSS‑TTS Family is an open‑source speech and sound generation model family from MOSI.AI and the OpenMOSS team. It is designed for high‑fidelity, high‑expressiveness, and complex real‑world scenarios, covering stable long‑form speech, multi‑speaker dialogue, voice/character design, environmental sound effects, and real‑time streaming TTS. | 3.3K | — | +161+5.1% | +1.5K+82.6% |
| 44 | PentesterFlow/agent Agentic offensive-security in your terminal | 450 | — | +166+58.5% | +437+3361.5% |
| 45 | axolotl-ai-cloud/axolotl Go ahead and axolotl questions | 12K | — | +31+0.3% | +153+1.3% |
| 46 | openai/symphony Symphony turns project work into isolated, autonomous implementation runs, allowing teams to manage work instead of supervising coding agents. | 25.3K | +1+0.0% | +194+0.8% | +1.8K+7.6% |
| 47 | Chainlit/chainlit Build Conversational AI in minutes ⚡️ | 12.2K | — | +33+0.3% | +141+1.2% |
| 48 | VectifyAI/PageIndex 📑 PageIndex: Document Index for Vectorless, Reasoning-based RAG | 32.9K | +1+0.0% | +305+0.9% | +1.9K+6.0% |
| 49 | git-ai-project/git-ai A Git extension for tracking the AI-generated code in your repos | 2.1K | — | +29+1.4% | +217+11.8% |
| 50 | adysec/clawbot iLink bot management tool with CLI and web dashboard. | 37 | +37 | — | — |