LLM: open-source projects on GitHub
2094 repositories on LLM, picked by the REDDYX AI agent. Each has a summary, key features and a breakdown. Sorted by stars.
// RELATED TOPICS
CODE-INTERPRETER - RUN AI CODE
A Python and JS/TS SDK package for securely executing AI-generated code directly in your application.
★ 2,333 · Pythonclaude-real-video - WATCH VIDEO WITH CLAUDE
Allows any LLM to "watch" video: extracts key frames and transcript, works locally.
★ 2,319 · PythonAGENTIC CONTEXT ENGINE
The platform enables agents to learn from experience and improve their responses. Now available as a hosting solution on kayba.ai.
★ 2,316 · PythonFABLE-METHOD: THREE KEYS
Protocol from Claude Fable 5: three steps – Think, Act, Prove – turned into a real workflow for any LLM.
★ 2,313 · PythonMAXTEXT JAX LLM
Simple, fast, and scalable framework for large language models on JAX and Python.
★ 2,309 · PythonSILLYTAVERN IN RUST
Classic Sillytavern rewritten in Tauri/Rust — faster, lighter, more reliable. 716 stars on GitHub.
★ 2,308 · JavaScriptFIGURES4PAPERS: IDEAL GRAPHS FOR PAPERS
Python scripts for creating quality graphs for publications in top AI conferences and journals.
★ 2,299 · PythonCode Agents Manager
Manage multiple Claude Code, OpenCode and other agents via TUI or web interface, with mobile access support.
★ 2,288 · RustTrainer: Distributed Training
The trainer project enables distributed training of AI models and fine‑tuning of LLMs in Kubernetes, written in Go.
★ 2,284 · GoCURATED RESOURCES FOR IN-CONTEXT LEARNING
Collection of best materials on in-context learning and prompt engineering. 2248 stars on GitHub. Jupyter notebooks with current techniques for ChatGPT, GPT-3, and FlanT5
★ 2,253 · Jupyter NotebookPROMA: Universal AI Agent
Proma combines Claude, Pi and any LLM in a single agent that works in Feishu chats and your scripts.
★ 2,246 · TypeScriptbeelzebub - AI-Driven Low-Code Virtualization
Beelzebub is a low-code framework for safely deceiving systems, using AI to virtualize environments.
★ 2,238 · GoClaude Code Plugins & Skills
425 plugins, 2810 skills, and 200 agents for Claude Code. An open marketplace with the ccpi CLI package manager.
★ 2,236 · PythonLOCAL AI MODELS IN RUST
Instant and controllable pre-trained AI models running locally. 2205 stars — project gaining popularity.
★ 2,236 · RustMEMMI-AGENT SHARED AI MEMORY
Personal AI agent and local memory hub for all AI assistants. Provides a shared, controllable context storage so each AI remembers the same user.
★ 2,217 · TypeScriptRUN LLM LOCALLY IN NODE.JS
Run AI models directly on your computer via Node.js. Enforced JSON schema validation right at the generation level.
★ 2,205 · TypeScriptAwesome LLM Long Context - Top Papers
A curated selection of must-read articles and blogs on long context in LLMs: new methods, reviews, and practices.
★ 2,192Fast Local AI Engine
Rapid-MLX is the fastest local AI engine for Apple Silicon, running 4.2× faster than Ollama and providing a cached first-token time of just 0.08s.
★ 2,185 · PythonNESTIA - AI ASSISTANT FOR NESTJS
Nestia simplifies NestJS application development by providing ready‑to‑use utilities and integrating an AI chatbot for rapid prototyping.
★ 2,182 · TypeScriptAI CHATBOT FRAMEWORK ON PYTHON
Framework for creating chatbots with NLU and AI on Python. Supports LLM integration and provides a ready-made dialog manager. Actively developed, already 2166 stars on Gi
★ 2,182 · TypeScriptZEROSHOT - AUTONOMOUS DEVELOPER TEAM
Specify a task — get production-ready code. Supports Claude Code, OpenAI Codex, OpenCode, and Gemini CLI.
★ 2,174 · JavaScriptWAKU-AGENT — PERSONAL AI AGENT
A personal AI agent in Python: local launch, memory, task loop and evaluation — everything ready in just a few minutes of reading the code.
★ 2,173 · PythonTENSORFOLD: FAST LLM DECODING ON MAC.
TensorFold provides exact and fast LLM decoding on Apple Silicon via MLX, offering an OpenAI-compatible API. The Python project has already earned 526 stars and supports
★ 2,172 · PythonKERNEL-MEMORY: AI MEMORY
An open-source research project in C# that provides a flexible memory system for users, teams, and applications.
★ 2,162 · C#VLLM-ASCEND: ASCEND PLUGIN ON GITHUB
Community plugin for vLLM on Ascend hardware from the community, written in C++. Gaining stars — already 2156!
★ 2,156 · C++ZERO-TO-SGLANG: LLM INFERENCE TRAINING
The Datawhale and SGLang course teaches LLM inference from scratch: build a mini‑SGLang, study the source code, and land your first PR. Available in English and Chinese.
★ 2,136 · PythonLLM-Space — Prototype Agents Easily
Desktop application for developing and testing LLM agents with deep control of each step.
★ 2,130 · TypeScriptUNIFIED MCP SERVER MANAGER
Unified application for managing MCP servers. Start, configure, and orchestrate everything from one place.
★ 2,127 · TypeScriptVOICE AI ON ARDUINO
Realtime Voice AI with 100+ models on ESP32. Secure WebSocket and Edge Functions for AI devices.
★ 2,118 · TypeScriptMCP SERVER FOR GITLAB
MCP server for GitLab provides over 100 tools for working with projects, merge requests, issues, pipelines and wikis via REST and GraphQL API. Allows integrating AI agent
★ 2,105 · TypeScriptOPEN SCIENCE AGENTIC WORKSPACE
Open Science — a local AI workstation for researchers, an alternative to Claude Science supporting models and reproducible experiments.
★ 2,100 · TypeScriptAI LEGION FOR AUTONOMOUS PENTESTING
Autonomous pentest legion with AI commander and worker swarm. Tencent IAD champion: 3rd of 634 teams, 8h unattended, 1B tokens beat 4.5B. Implements Nüwa Test: gives AIs
★ 2,088 · PythonSOLID-GLASS DESIGN SYSTEM FOR AI
Solid‑glass — это TypeScript‑based дизайн‑система для создания интерфейсов AI‑агентов. Она предоставляет Material‑like компоненты, тактильную обратную связь и технические
★ 2,071 · TypeScriptSuperglue AI Integrator
Superglue turns natural language into ready tools and integrations for long-tail and enterprise systems. Get no-code solutions.
★ 2,069 · TypeScriptNOTTE - Framework for Web Agents
notte is a framework for web agents: create and deploy serverless automation functions in a reliable browser environment.
★ 2,061 · PythonFAST LLM LAUNCH ON NPU
FastFlowLM runs LLMs on AMD Ryzen AI NPU in minutes, delivering performance comparable to Ollama but optimized for NPU.
★ 2,050 · C++RUN NEURAL NETWORKS LOCALLY
A collection of the best platforms and tools for running LLM on your own hardware. 2000+ stars!
★ 2,045INDEX‑TRANSLATE MULTILINGUAL AI
Index‑Translate — a multilingual family of translation models based on LLM. Implemented in Python, 345★ on GitHub, supports fine‑tuning and inference for dozens of langua
★ 2,044 · PythonAWESOME LANGGRAPH: EVERYTHING UNDER MEASURE
Complete catalog of LangChain+LangGraph: QT concepts, projects, tools, and templates for LLM and multi-agent systems.
★ 2,043 · JavaScriptLLAMA.VIM: LLM RIGHT IN VIM
A Vim plugin that adds code and text completion using large language models. 2000+ stars on GitHub.
★ 2,042 · Vim ScriptKVCACHED: ACCELERATE GPU CACHE
Virtualized elastic KV‑cache for dynamic GPU sharing, accelerating LLM inference and reducing latency.
★ 2,041 · PythonSUPERLINKED INFERENCE ENGINE
Open-source server for embedding inference, reranking, and data extraction. 2000+ stars on GitHub.
★ 2,029 · PythonAI-ASSISTANT PY-GPT FOR EVERYONE
Desktop AI Assistant based on multiple models: GPT-5, GPT-4, Gemini and others, with voice, vision, RAG, and image generation.
★ 2,027 · PythonRAGHUB: Resources and RAG Frameworks
The community gathers the best RAG frameworks, projects, and materials for AI developers. Contribute and discover new solutions.
★ 2,019ANY-LLM: One Interface for All LLMs
Python library for unified interaction with LLM providers through a single interface.
★ 2,010 · PythonLOCAL OPEN-SOURCE AI AGENT
Atomic-Chat is a private AI engine for running LLM locally without internet, fully offline on your computer. ⭐ 967 stars on GitHub.
★ 2,008 · TypeScriptVLLM-METAL: ACCELERATING LLM ON APPLE SILICON
vllm-metal is a community‑maintained plugin that brings Metal‑based acceleration to vLLM on Apple Silicon. It uses MLX to run LLMs on macOS with low latency and high thro
★ 1,997 · PythonOBSIDIAN AI TEXT GENERATOR
The plugin lets you generate text in Obsidian using OpenAI, Anthropic, Google, and local models.
★ 1,996 · TypeScriptMCP - Agent with MS Docs
Official Microsoft Learn MCP server and CLI — gives AI agents instant access to verified documentation and code samples.
★ 1,992 · TypeScriptLLM-Native 3D Modeling Library
Python library letting LLMs create, edit, and export 3D models through a code-centric API.
★ 1,992 · HTMLONE FILE, ALL DATA
A utility that gathers content from GitHub, arXiv, YouTube, etc. into a single text file for feeding into LLM.
★ 1,977 · PythonCOLLECTION OF JAILBREAK LLM METHODS
Awesome-Jailbreak-on-LLMs is a collection of cutting‑edge jailbreak methods for large language models. The repository contains papers, code, datasets, evaluations, and ef
★ 1,977Nanocoder — Local Assistant
Beautiful local coding agent for the terminal, built by the community for the community ⚒
★ 1,957 · TypeScriptNEURON-AI PHP FRAMEWORK
A PHP framework for building AI agents that connects LLM, vector databases, and memory.
★ 1,951 · PHPHELLOAGENTS - Easy Agent Start
A simple framework for creating AI agents in Python, based on the hello-agents tutorial. Allows you to quickly build and run autonomous assistants.
★ 1,945 · PythonBRUTEFORCEAI - SMART BRUTE
A Python tool that uses LLM for smart password guessing and automating login attacks.
★ 1,935 · PythonMCP SERVER FOR CODE CONTEXT
aoci-code — local MCP server and CLI in Go. Indexes code and DB schemas into a unified repo map. Helps AI agents (Claude Code, Cursor, Codex) understand complex systems f
★ 1,918 · GoAI Copilot for Blender
Connects Blender to LLMs via MCP for visual chat, scene manipulation, and multi-agent 3D workflows.
★ 1,918Agent-Field: AI Agents as Services
Create, launch, and scale AI agents as API microservices with observability, audit, and identity from day one.
★ 1,911 · GoGRAPHS VS FRAUD
Curated list of articles on fraud detection using graphs and transformers.
★ 1,907New projects on this topic in Telegram
Telegram →