Open-source large-language-models software
3737 open-source libraries, tools and applications tagged large-language-models, ranked by how closely large-language-models is their primary purpose, then by adoption and maintenance health. Each entry links to a page with sources, use cases and when to choose or avoid it.
3737 products, primary matches first, then adoption-weighted; health v2 shown.
| Product | Health v2 | Stars | Maturity |
|---|---|---|---|
| lucidrains/flamingo-pytorch A PyTorch implementation of DeepMind's Flamingo visual language model architecture, providing the Perceiver Resampler and Gated Cross-Atten… | 23 | 1267 | active |
| facebookresearch/DrQA DrQA is a PyTorch implementation of a system for open-domain question answering that combines document retrieval over Wikipedia with a neur… | 10 | 4467 | maintenance |
| yym68686/uni-api uni-api is a self-hosted Python service that unifies management of multiple LLM provider APIs behind a single OpenAI-compatible endpoint. I… | 84 | 1266 | active |
| TencentCloudADP/youtu-graphrag Youtu-GraphRAG is a Python framework for graph-based retrieval-augmented generation that unifies graph schema construction, community detec… | 52 | 1266 | active |
| Storia-AI/sage Sage is an open-source tool that lets developers chat with any codebase to quickly understand how it works, similar to a self-hosted GitHub… | 10 | 1265 | active |
| NVIDIA-NeMo/Gym NeMo Gym is a Python library from NVIDIA for evaluating and improving LLM models and agents using environments. It provides infrastructure … | 81 | 1264 | active |
| SimonSchubert/Kai Kai 9000 is an open-source, cross-platform AI assistant built with Kotlin Multiplatform that runs on Android, iOS, Windows, Mac, Linux, and… | 81 | 1264 | active |
| Arize Phoenix OpenInference is a set of OpenTelemetry-based conventions and instrumentation plugins for tracing AI and LLM applications. It provides pack… | 90 | 1263 | active |
| techjarves/Uncensored-Local-AI-Multiplatform A cross-platform Flutter application that runs uncensored open-source GGUF language models entirely on-device for Android, iOS, and desktop… | 62 | 1263 | active |
| meta-pytorch/attention-gym Attention Gym is a collection of tools, examples, and reference implementations for working with PyTorch's FlexAttention API. It provides a… | 91 | 1260 | active |
| Taxuspt/garmin_mcp A Model Context Protocol (MCP) server that connects to Garmin Connect and exposes fitness and health data (activities, sleep, heart rate, t… | 67 | 1260 | active |
| amElnagdy/guard-skills A collection of guard skills for AI coding agents that act as second-pass quality gates on AI-generated code, tests, and documentation. It … | 51 | 1260 | active |
| HelgeSverre/ollama-gui Ollama GUI is a web-based chat interface for interacting with local LLMs served through the Ollama API. It offers local chat history via In… | 95 | 1258 | active |
| CopilotKit/OpenTag OpenTag is an open-source, self-hosted knowledge-work AI agent for Slack and Microsoft Teams, built as the starter application for the Chan… | 78 | 1258 | active |
| plurai-ai/intellagent IntellAgent is a Python framework for diagnosing and optimizing conversational LLM agents through simulated, realistic synthetic user inter… | 55 | 1258 | active |
| thu-ml/SpargeAttn SpargeAttn is the official implementation of a training-free sparse and quantized attention method that accelerates inference for language,… | 48 | 1257 | active |
| hkjarral/AVA-AI-Voice-Agent-for-Asterisk An open-source AI voice agent that integrates with Asterisk/FreePBX phone systems via Audiosocket/RTP, built in Python with a modular pipel… | 85 | 1256 | active |
| MoonshotAI/FlashKDA FlashKDA is a set of high-performance CUDA kernels (built on CUTLASS) implementing Kimi Delta Attention, a linear attention mechanism, for … | 56 | 1256 | active |
| ENTERPILOT/GoModel GoModel is an open-source AI gateway and control plane written in Go that exposes a unified OpenAI-compatible and Anthropic-compatible API … | 80 | 1255 | active |
| grinev/opencode-telegram-bot A Telegram bot that acts as a secure mobile client for the OpenCode CLI, letting users run and monitor AI coding tasks from their phone whi… | 79 | 1254 | active |
| TIGER-AI-Lab/OpenResearcher OpenResearcher is a fully open-source pipeline for synthesizing long-horizon deep research trajectories using LLM agents with retrieval and… | 51 | 1254 | active |
| myreader-io/myGPTReader myGPTReader is a Slack bot powered by ChatGPT that reads and summarizes webpages, documents (eBooks, PDF, DOCX), and YouTube videos, and su… | 10 | 4414 | maintenance |
| facebookresearch/llm-transparency-tool An interactive toolkit from Meta Research for analyzing the internal workings of Transformer-based language models. It visualizes contribut… | 10 | 1253 | active |
| EmergenceAI/Agent-E Agent-E is an open-source agent-based system that automates actions on the user's computer, currently focusing on browser automation via na… | 58 | 1252 | active |
| stephenschoettler/hermes-lcm A lossless context management plugin for Hermes Agent that replaces one-shot prompt compression with a SQLite-backed, DAG-based context eng… | 76 | 1251 | active |
| ChristopherKahler/paul PAUL (Plan-Apply-Unify Loop) is a structured development framework for Claude Code that enforces a PLAN → APPLY → UNIFY cycle with acceptan… | 57 | 1251 | active |
| smixs/pohuy A prompt/skill package that makes AI agents respond in idiomatic Russian profanity, compressing verbose explanations into terse, expressive… | 75 | 1250 | active |
| zsyggg/paper-craft-skills A collection of Claude Code / Codex agent skills that turn academic papers into method figures, visual slide decks, and in-depth HTML artic… | 50 | 1250 | active |
| RUC-NLPIR/Search-o1 Search-o1 is a research framework that enhances large reasoning models (like QwQ and R1) with agentic search and retrieval-augmented genera… | 41 | 1247 | active |
| openai/codex-action A GitHub Action that runs OpenAI's Codex CLI inside GitHub Actions workflows, installing the CLI and routing API calls through a secure pro… | 60 | 1246 | active |
| firecrawl/web-agent An open-source TypeScript framework for building autonomous web research agents, layered from a Next.js chat template down to an agent core… | 47 | 1246 | active |
| MaoXiaoYuZ/Long-Novel-GPT Long-Novel-GPT is a self-hosted web application that uses LLMs and RAG to generate and revise long-form novels through a top-down outline-c… | 46 | 1246 | active |
| xllm-go/bypass A Go server that reverse-engineers the chat interfaces of multiple AI providers (Coze, DeepSeek, Cursor, Windsurf, Grok, Bing Copilot, You,… | 56 | 1245 | active |
| mrwadams/attackgen AttackGen is a Streamlit-based cybersecurity tool that uses large language models to generate tailored incident response testing scenarios … | 91 | 1244 | active |
| RinDig/Interpretable-Context-Methodology Interpretable Context Methodology (ICM) is a Python-based methodology that uses folder structure as agent architecture, replacing framework… | 54 | 1244 | active |
| sums001/Windows-Copilot-API A Python library and local server that reverse engineers Microsoft Copilot's free web chat into an OpenAI-compatible REST API, giving acces… | 50 | 1244 | active |
| HJYao00/Mulberry Mulberry is a research implementation of an o1-like multimodal large language model (MLLM) that performs step-by-step reasoning and reflect… | 46 | 1244 | active |
| Ksuriuri/index-tts-vllm A reimplementation of IndexTTS's GPT model inference using vLLM, providing significantly faster text-to-speech generation with a web UI and… | 53 | 1243 | active |
| AWorld AWorld is an open-source agent framework and runtime that orchestrates tools, memory, context, and execution for building autonomous AI age… | 76 | 1242 | active |
| aegra/aegra Aegra is an open-source, self-hosted backend server for running LangGraph AI agents, serving as a drop-in replacement for LangSmith Deploym… | 84 | 1241 | active |
| Taewan-P/gpt_mobile GPT Mobile is an Android chat application that lets users talk to multiple LLM providers (OpenAI, Anthropic, Google Gemini, Groq, Ollama) s… | 96 | 1240 | active |
| daaain/claude-code-log A Python CLI tool that converts Claude Code transcript JSONL files into readable HTML and Markdown logs, with an interactive TUI for browsi… | 85 | 1240 | active |
| 0xacx/chatGPT-shell-cli A lightweight shell script that lets you chat with OpenAI's ChatGPT models and generate DALL-E images directly from the terminal, requiring… | 32 | 1240 | active |
| Shopify/roast Roast is a Ruby-based domain-specific language for building structured AI workflows from composable building blocks called cogs. It orchest… | 80 | 1239 | active |
| thClaws/thClaws thClaws is an open-source AI agent harness written in native Rust that ships as a single binary offering a desktop GUI, CLI, headless, and … | 78 | 1239 | active |
| supavec/supavec Supavec is an open-source RAG-as-a-Service platform (an alternative to Carbon.ai) that lets developers upload documents, generate embedding… | 45 | 1239 | active |
| lcoutodemos/clui-cc Clui CC is a macOS desktop overlay application that wraps the Claude Code CLI in a floating, transparent pill interface with multi-tab sess… | 45 | 1239 | active |
| GML-MMGroup/GMTalker GMTalker is an interactive 3D digital human system rendered with Unreal Engine, integrating speech recognition, speech synthesis, natural l… | 42 | 1239 | active |
| TheSmallHanCat/sora2api A self-hosted OpenAI-compatible API gateway that wraps Sora's text-to-video and image generation capabilities behind standard /v1/chat/comp… | 10 | 1235 | active |
| laravel/ai The Laravel AI SDK is a PHP package offering a unified, expressive API for interacting with AI providers such as OpenAI, Anthropic, and Gem… | 84 | 1234 | active |
| modu-ai/moai-adk MoAI-ADK is an agentic development harness for Claude Code that enforces SPEC-driven plan/run/sync workflows, TRUST 5 quality gates, and mu… | 78 | 1234 | active |
| InternScience/GraphGen GraphGen is a Python framework for knowledge-graph-guided synthetic data generation for LLM training. It builds fine-grained knowledge grap… | 53 | 1232 | active |
| mikiarlo3/ai-copywriter A portable Markdown-based agent skill that writes marketing copy (headlines, descriptions, microcopy, subject lines) with a human tone whil… | 53 | 1232 | active |
| Scale3-Labs/langtrace Langtrace is an open-source, OpenTelemetry-based observability platform for LLM applications, providing real-time tracing, metrics, and eva… | 47 | 1232 | active |
| OpenTSLM/OpenTSLM OpenTSLM is a family of Time-Series Language Models that integrate time series as a native modality into pretrained LLMs (Llama, Gemma), en… | 58 | 1230 | active |
| bricks-cloud/BricksLLM BricksLLM is a cloud-native AI gateway written in Go that sits between applications and LLM providers like OpenAI, Anthropic, Azure OpenAI,… | 31 | 1230 | active |
| wanxingai/LightAgent LightAgent is an ultra-lightweight open-source Python framework for building LLM-powered agents with tools, persistent memory, guardrails, … | 87 | 1229 | active |
| sentient-agi/EvoSkill EvoSkill is an open-source Python framework that automatically discovers and synthesizes reusable skills from failed agent trajectories to … | 72 | 1229 | active |
| microsoft/MInference MInference is a Microsoft library that accelerates long-context LLM inference using dynamic sparse attention, reducing pre-fill latency by … | 60 | 1229 | active |
| leezythu/maoxuan-skill A Claude Code skill package that distills the cognitive frameworks from Mao Zedong's Selected Works (e.g., contradiction analysis, united f… | 56 | 1229 | active |
| 666ghj/DeepSearchAgent-Demo A framework-free Python implementation of a deep search AI agent that generates high-quality research reports through multi-round web searc… | 31 | 1229 | active |
| jx-sec/jxwaf JXWAF is an open-source web application firewall powered by an AI large language model, combining an AI security model, a semantic analysis… | 68 | 1228 | active |
| sgl-project/SpecForge SpecForge is a Python framework from the SGLang team for training speculative decoding models such as EAGLE/EAGLE3 draft heads. Trained mod… | 64 | 1228 | active |
| metoro-io/mcp-golang mcp-golang is an unofficial Go implementation of the Model Context Protocol (MCP) for building MCP servers and clients with minimal boilerp… | 60 | 1228 | active |
| langchain-ai/openevals OpenEvals is a library of readymade evaluators for testing LLM applications, including LLM-as-judge evaluators and reference-based/referenc… | 89 | 1227 | active |
| uezo/ChatdollKit ChatdollKit is a Unity SDK that turns 3D character models into voice-enabled chatbots and virtual assistants. It integrates LLMs (ChatGPT, … | 84 | 1227 | active |
| mem9-ai/mem9 mem9 is a persistent memory layer for AI agents, providing cross-session and cross-machine memory, shared memory for multi-agent workflows,… | 60 | 1227 | active |
| MoonshotAI/Kimi-VL Kimi-VL is an open-source Mixture-of-Experts vision-language model (VLM) with a 2.8B activated parameter language decoder, offering multimo… | 31 | 1227 | active |
| Bolin97/GongBU GongBU is a self-hosted, no-code web platform for fine-tuning, evaluating, and deploying large language models, built on Transformers and P… | 49 | 1225 | active |
| Opencode-DCP/opencode-dynamic-context-pruning A plugin for OpenCode that dynamically prunes and compresses conversation context to reduce token usage when working with LLMs. It replaces… | 77 | 4294 | maintenance |
| Truss Truss is a Python CLI and packaging framework for deploying and serving AI/ML models in production, primarily on the Baseten platform. It h… | 96 | 1223 | active |
| GoogleCloudPlatform/vertex-ai-creative-studio GenMedia Creative Studio is a web application showcasing Google Cloud's generative media APIs including Gemini Image (Nano Banana), Veo vid… | 87 | 1223 | active |
| FlowDown FlowDown is a native AI chat client for iPhone, iPad, and Mac built with Swift and UIKit. It connects to OpenAI-compatible APIs, self-hoste… | 85 | 1223 | active |
| allwefantasy/auto-coder Auto-Coder is an open-source AI coding agent and CLI tool (powered by Byzer-LLM) that provides chat, one-shot command, server, and RAG mode… | 66 | 1221 | active |
| gabrielchua/RAGxplorer RAGxplorer is a Python library and hosted Streamlit app for visualizing Retrieval Augmented Generation pipelines. It chunks PDFs, embeds th… | 17 | 1221 | active |
| EvolvingLMMs-Lab/LLaVA-OneVision-2 A fully open framework for training multimodal large language models, releasing models, datasets, and training recipes for the LLaVA-OneVis… | 70 | 1220 | active |
| vincanger/coverlettergpt A full-stack web application that generates tailored cover letters from a job description and the user's resume using OpenAI's GPT models. … | 40 | 1220 | active |
| moltlaunch/cashclaw CashClaw is an autonomous AI agent that connects to the Moltlaunch onchain marketplace, where it evaluates tasks, quotes prices, executes w… | 44 | 1219 | active |
| AllAboutAI-YT/easy-local-rag A Python application providing 100% local retrieval-augmented generation (RAG) using Ollama for LLM inference and embeddings. It supports i… | 16 | 1219 | active |
| tonyqinatcmu/SlideBot-AI SlideBot AI is an AI-powered presentation generator that turns a topic, outline, or uploaded materials (documents, spreadsheets, meeting re… | 42 | 1216 | active |
| RhythmicWave/NovelForge NovelForge is an AI-assisted long-form novel writing application built in Python, using a card-based creation model with JSON Schema-driven… | 83 | 1215 | active |
| ZHangZHengEric/Sage Sage is a Python-based multi-agent framework and platform for planning, executing, and delivering complex tasks with LLM-powered agents. It… | 87 | 1214 | active |
| Symfony AI Symfony AI is a set of PHP components and Symfony bundles that integrate AI capabilities into PHP applications, including a unified interfa… | 85 | 1214 | active |
| Zefan-Cai/R-KV R-KV is a training-free, redundancy-aware KV cache compression method for reasoning LLMs, discarding repetitive tokens on-the-fly during de… | 64 | 1214 | active |
| ArcadeData/arcadedb ArcadeDB is an open-source, Apache 2.0 licensed multi-model DBMS written in Java that natively stores graphs, documents, key-value pairs, f… | 100 | 1213 | active |
| anthropics/anthropic-sdk-go The official Go SDK for accessing Anthropic's Claude language model APIs. It provides typed clients for messages, streaming, and other Clau… | 85 | 1212 | active |
| SalesforceAIResearch/enterprise-deep-research Enterprise Deep Research (EDR) is a multi-agent deep research system from Salesforce AI Research that combines a master planning agent, spe… | 52 | 1212 | active |
| Morgensonne/EditDeck EditDeck is an end-to-end pipeline that turns plain-language requirements into structured slide outlines, rendered slide images, standard P… | 47 | 1212 | active |
| NVIDIA/kvpress kvpress is a Python library from NVIDIA that implements multiple KV cache compression methods and benchmarks for long-context LLM inference… | 86 | 1211 | active |
| xenodium/chatgpt-shell An Emacs package providing a multi-LLM shell and chat interface supporting ChatGPT, Claude, DeepSeek, Gemini, Kagi, Ollama, and Perplexity,… | 74 | 1210 | active |
| modal-labs/quillman QuiLLMan is a voice chat application built on Kyutai's Moshi speech-to-speech language model, deployed serverlessly on Modal with a FastAPI… | 65 | 1210 | active |
| jieyefriic/rp-engine A YAML-native agent workflow execution engine written in Rust that parses declarative workflow files describing nodes, edges, prompts, data… | 61 | 1209 | active |
| Elixir LangChain An Elixir framework for building LLM-powered applications, providing abstractions and integrations for chat models like OpenAI, Anthropic C… | 95 | 1208 | active |
| letta-ai/agent-file Agent File (.af) is an open file format for serializing stateful AI agents, packaging system prompts, persistent memory, tool configuration… | 65 | 1208 | active |
| cvs-health/uqlm UQLM is a Python library for detecting LLM hallucinations using uncertainty quantification techniques. It provides five categories of score… | 86 | 1207 | active |
| FunnyWolf/agentic-soc-platform Agentic SOC Platform (ASP) is an open-source, agent-centric security operations platform that ingests SIEM/webhook alerts, correlates them … | 83 | 1207 | active |
| clacky-ai/openclacky OpenClacky is an open-source (MIT) AI agent platform written in Ruby that runs locally, offering a CLI and Web UI for completing tasks via … | 80 | 1207 | active |
| superlinear-ai/raglite RAGLite is a Python toolkit for building Retrieval-Augmented Generation (RAG) pipelines backed by DuckDB or PostgreSQL, with hybrid keyword… | 80 | 1207 | active |
| microsoft/RecAI RecAI is a Microsoft research project exploring ways to integrate large language models into recommender systems (LLM4Rec). It includes a c… | 54 | 1205 | active |
| googleapis/go-genai The official Google Gen AI SDK for Go, providing a client interface to Google's generative models such as Gemini via the Gemini Developer A… | 85 | 1204 | active |