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
CONTEXT AGENTS KIT
A set of skills for Claude Code that improve the quality of AI agents' work. Compatible with OpenCode, Cursor, Gemini CLI, and others.
★ 1,075 · TypeScriptWISP-SCIENCE: LOCAL AI WORKSTATION
Open-source local workstation for scientific computing with Python/R, bioinformatics, SSH/WSL/GPU and OpenAI/Anthropic models.
★ 1,075 · HTMLLLAMA NOW IN REACT NATIVE
llama.rn is a React Native binding for llama.cpp, enabling LLM models to run directly on mobile devices. 981 stars on GitHub.
★ 1,074 · C++ASREVIEW - ACTIVE LEARNING FOR REVIEWS
Asreview uses active learning to speed up systematic literature reviews, reducing the number of required readings.
★ 1,074 · PythonRunning Laya in Node.js
The project provides TypeScript bindings to run the open Laya decision model, compatible with Jev System-1, via ONNX Runtime in Node.js.
★ 1,074 · TypeScriptREDUCE AI TOKEN COSTS
Chisle cuts AI-agent token usage: terse prose, YAGNI-first code, and tool-output compression. Works with Claude Code, Pi, Cursor, Codex, Gemini and four more, zero deps,
★ 1,072 · JavaScriptOPENCODE MANAGER AI AGENTS
OpenCode web‑center unifies sessions, git, terminal and schedules in a single interface. Accessible from phone and desktop, installed with one Docker command.
★ 1,072 · TypeScriptANYMD: CONVERT FILES TO MARKDOWN!!
AnyMMD converts PDF, Word, PowerPoint, Excel, EPUB, HTML and images to clean Markdown for AI agents. Works locally on Rust MCP server + CLI, no API key needed. Fast extra
★ 1,070 · TypeScriptAI-NATIVE FILE SYSTEM IN RUST
Curvine is an AI-native file system in Rust with full POSIX on object storage. Features multi-tier distributed cache for training and inference. CNCF Sandbox project.
★ 1,068 · RustSEARCHAT AI SEARCH CHAT
SearChat combines AI chat and search in a single interface. Supports OpenAI, Anthropic, VertexAI, Gemini, DeepResearch and the SearXNG aggregator. Available as a Docker i
★ 1,064 · TypeScriptAI Automation on Java
Open platform for orchestrating AI agents and automating workflows. Combines autonomy and precision in one solution.
★ 1,063 · JavaFRAMEWORK FINE-TUNING FOR iOS 26
Examples of apps for the Foundation Models Framework on iOS 26 and macOS 26 — exploring fine‑tuning of LLMs directly on the device.
★ 1,060 · SwiftCodeSight: AI Context
Universal context generator for AI assistants. Saves thousands of tokens in Claude Code, Cursor, Copilot, and others.
★ 1,059 · TypeScriptAGENTS-FLEX AI-Agent Framework
Lightweight Java framework for building AI agents with support for RAG, MCP, Skills, and Text2SQL.
★ 1,056 · JavaCLIRELAY - UNIFIED GATEWAY FOR AI CODING
Self-hosted AI gateway unifies OpenAI, Claude, Gemini, and Codex in a single endpoint with a web console, request logs, and cost quotas.
★ 1,055 · GoRAFT - Accelerate ML on CUDA
The RAFT library provides CUDA‑accelerated ML and IR algorithms, simplifying the creation of high‑performance applications.
★ 1,054 · CudaMASSGEN - Multi-Agent System
MassGen — an open multi-agent system for the terminal that coordinates models and agents to produce high-quality results.
★ 1,052 · PythonDYNAMIQ: AI APPLICATION ORCHESTRATION
Dynamiq is a framework for orchestrating agentic AI and LLM applications in Python. A simple way to create scalable AI workflows.
★ 1,052 · PythonMULTIMODAL FUTURE IS HERE
Fully open-source framework for training multimodal models — accessible to everyone. Python.
★ 1,050 · PythonTEXTFOLD: MULTILEVEL TEXT COMPRESSIONS.
Textfold lets you collapse scientific articles, web pages, files, and AI replies into five levels of detail — from a high‑level overview to the original text. The Python
★ 1,046 · PythonTENSORLAKE: SERVERLESS FOR AGENTS 945★
Serverless runtime for sandboxes and launching background agent applications. Python, stars are growing — see for yourself!
★ 1,044 · PythonOPENLOOMI: YOUR AI WORK MATE
Openloomi is an open AI assistant that remembers all details of your work and offers helpful hints in real time.
★ 1,043 · TypeScriptSECURE WITHOUT CONTAINERS
secure-exec — safe execution of Node.js without sandbox. Lightweight library without containers and VMs, compatible with npm out of the box.
★ 1,041 · JavaScriptCSGHUB-SERVER - MANAGE DATA AND MODELS
Go‑based CSGHub backend for managing datasets, models, running inference, fine‑tuning, and deploying applications.
★ 1,041 · GoAI Agent Evaluation Toolkit
A curated library of top resources for building and evaluating AI agents, including articles, blogs, tools, and benchmarks.
★ 1,039RUN LARGE LLMS LOCALLY
PolyStrata runs Qwen3.8-Flash, GLM 5.3 Flash, DeepSeek V4 and more on your PC. One-click install for Windows and Linux with OpenAI/Anthropic API on localhost.
★ 1,037 · C++ZEROSTACK - Rust Agent for Code
A minimalist AI agent for programming in Rust. Optimized for memory and speed — already 1036 stars on GitHub!
★ 1,036 · RustJUDGEVAL STACK FOR AGENTS
Platform for continuous improvement of AI agents: collecting environment data and evaluations for monitoring and growth.
★ 1,036 · PythonGraphRAG-SDK - CREATE FAST GENAI
GraphRAG-SDK lets you quickly build accurate GenAI applications at scale, using graph RAG and advanced retrieval.
★ 1,036 · PythonJapanese NLP: Top Resources
A collection of libraries, models, corpora, and datasets for Japanese language processing in Python. Includes Claude Code skills for quick resource discovery.
★ 1,031Distributed LLM for All
A decentralized network for sharing computing resources and running AI agents and chatbots.
★ 1,030 · RustEASY-AGENT: AI CODING AGENT IN TERMINAL
Production‑ready open‑source terminal agent in TypeScript with access rules, OS sandboxing, MCP, skills and sub‑agents. Supports Anthropic, OpenAI‑compatible, Gemini and
★ 1,029 · TypeScriptTUTEL ACCELERATES MOE IN C
Optimized Mixture-of-Experts library for accelerated LLM inference with FP8, NVFP4, and MXFP4 support.
★ 1,028 · CAI Testing with alumnium
alumnium is an AI‑driven end‑to‑end testing framework that generates test scripts and automatically adapts them to UI changes.
★ 1,024 · TypeScriptSiliconScope — Apple Silicon Monitor
Native Apple Silicon system monitor in SwiftUI without sudo: tracks ANE, Media Engine, and memory bandwidth.
★ 1,024 · SwiftMCPM.SH CLI MANAGER MCP
MCPM.SH — cross‑platform CLI package manager for MCP with a registry for all clients. It allows searching and configuring MCP servers via a smart router and profiles.
★ 1,023 · PythonTPU-MLIR: ACCELERATING AI ON TPU
MLIR compiler for accelerating machine learning models on Sophgo TPU chips. Open source C++ code with 952 stars.
★ 1,022 · C++InferenceX – Inference Benchmark
Open-source continuous inference benchmark for Qwen3.5, DeepSeek, GPTOSS and new accelerators.
★ 1,016 · PythonCODEAGENT: AI AGENT FOR TERMINAL
CodeAgent — open-source AI agent working in the terminal. It analyzes the repo, plans changes, edits files, runs tests and repeats the loop until the task is done.
★ 1,012 · PythonDSCODE AI AGENT CODING
DSCODE provides a persistent shell with Ultra subagents, auto‑approval of changes and Chrome MCP integration. The system collects session telemetry to analyse coding‑agen
★ 1,001 · JavaScriptKeras Hub: Models for Keras
Platform with ready-to-use pretrained models for Keras 3, simplifying AI integration into projects.
★ 999 · PythonHAX MINIMALIST TERMINAL AGENT
Minimalist coding agent in C, running directly in the terminal. Allows writing and editing code without extra dependencies.
★ 999 · COPEN INTELLIGENT CHAT UI
Open-intelligent-ui — open-source implementation of ChatGPT's intelligent UI: answers stream as interactive maps, photos, charts, drawings and working tools. Written in T
★ 997 · TypeScriptCALI - AI Agent React Native
Create React Native apps in minutes with the cali AI agent. Automate UI, navigation, and state.
★ 992 · TypeScriptLLM Architecture
A step‑by‑step exploration of LLMs, from tokenization through attention to inference optimization.
★ 989How to Train Your LLM
Build a modern LLM from scratch — every line of code is commented and explained simply, as if for a five-year-old.
★ 984 · Jupyter NotebookLOCAL AI‑TEAMMATE FOR YOUR PROJECTS
Please, Local AI‑teammate workspace Thank you. Please Connect your models and keep project dialogues, tool approvals, and results together in one place Thank you.
★ 981 · TypeScriptRust Framework for Scalable Workflows
Flow-Like: Strongly Typed Enterprise-Scale Workflows. Built for scalability, speed, seamless AI integration, and rich customization.
★ 980 · RustAI Metasearch & Grounding Engine
Self-hosted metasearch and grounding engine for AI agents, built on SearXNG with JSON API and optional page text extraction.
★ 978 · RustMASTERAGENT - POWERFUL OFFLINE AI AGENT
MasterAgent enables creating AI agents that run fully on‑device with sub‑100ms latency on Qualcomm NPU, requiring no cloud.
★ 976 · C++DETERMINFLOW: AI WORKFLOWS
Platform for reliable creation, testing, and deployment of complex AI workflows as services.
★ 975 · PythonSelf-Hosted Web Search Engine
Self-hosted, SSRF-hardened web search and grounding engine for local AI agents. Built in Rust with AGPL-3.0-or-later license.
★ 973 · RustONE ID FOR AI AGENT
Unified on_message handler for Slack, Discord, Telegram and other channels. Python and TypeScript SDK + CLI.
★ 972 · PythonOPENEVEREST - AUTOMATE DB IN K8S
OpenEverest is an open platform for automatic provisioning and management of databases on any Kubernetes infrastructure.
★ 972 · TypeScriptAI Application Evaluation
Open platform for evaluating and improving LLM and AI agents with tracing, simulation, and dataset support. Runs locally.
★ 971 · PythonAWESOME-HOSTING TOP CHEAP HOSTINGS
A curated selection of the best hostings, sorted by the price of the minimal plan. Quickly find the most cost-effective option for your project.
★ 970DATA PREP KIT: ACCELERATE AI GENERATION
Open-source toolkit for preparing data to train and use GenAI models. Simplifies cleaning, transformation, and labeling of datasets.
★ 969 · HTMLAgileRL - Evolutionary Acceleration
AgileRL speeds up RL training 10× via evolutionary hyperparameter optimization and ready-to-use SOTA algorithms.
★ 967 · PythonSELFHOST AI: ONE COMMAND
Deploy n8n, Ollama, Flowise, RAG and over 30 tools with a single command, including automatic HTTPS.
★ 963 · ShellJEVOS: BINARY DECISIONS ON LAPTOP
Jevos is an open-source Jev alternative for yes/no decisions. Runs fully locally on CPU via FastAPI. Data never leaves your device.
★ 963 · PythonNew projects on this topic in Telegram
Telegram →