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
AI ASSISTANT FOR EMACS AIDER
The aider.el package integrates the Aider AI assistant into Emacs, allowing you to generate, edit, and explain code directly from the buffer. It supports local models via
★ 688 · Emacs LispOPENTYPELESS - Voice typing everywhere
Opentypeless lets you type text by voice in any environment: speak, record, polish the result, and get ready text.
★ 684 · TypeScriptRANKIFY: TOOL FOR RAG
Rankify — Python toolkit for search, reranking, and RAG. Includes 40 benchmark datasets, 7+ search methods and 24+ ranking models. Supports multiple RAG schemes and easy
★ 683 · PythonFENIC: SEMANTIC DATAFRAME FOR AGENTS
Fenic is a Python library that adds semantics to DataFrames, making them understandable to humans and AI agents.
★ 681 · PythonBIOMCP: Biomedical AI Protocol
BioMCP is an open Rust protocol for integrating biomedical models into AI agents. It enables fast data exchange and LLM compatibility.
★ 680 · RustBest OpenClaw Resources
A curated list of the best OpenClaw resources: official projects, plugins, dashboards, and instructions.
★ 679DIFY-HELM: QUICK LLM DEPLOY
Helm chart for deploying Dify — an LLM application from langgenious in Kubernetes in minutes.
★ 679 · Go TemplateTENSORSHARP - POWERFUL .NET LLM
A library for quickly launching GGUF models on .NET with GPU support, console app, web chat, and an API compatible with Ollama and OpenAI.
★ 679 · C#WCGW - MCP Agent
Python agent for MCP clients: runs shell commands, provides code autocompletion, and enables quick script integration.
★ 679 · PythonSQZ: COMPRESSION FOR AI AGENTS
SQZ compresses tool output before it reaches the LLM, replacing repeated fragments with short 13‑token references. Works with Claude Code, Cursor, Codex, Kiro, Zed and an
★ 677 · RustSLOTSTREAM LAUNCHES QWEN3.8 ON MAC
Run Qwen3.8‑Flash‑Next (125B MoE, 104 GB at 4‑bit) on Mac by streaming experts from SSD on demand. Implemented with MLX + Swift, it offers an Ollama‑compatible API.
★ 673 · SwiftPYTHON-UTCP: AI AGENTS WITHOUT MIDDLEWARE
Official Python implementation of UTCP — an open standard allowing AI agents to directly call any API without middleware.
★ 672 · PythonFLUTTER RULES FOR AI CODERS
A set of AI rules and skills for Flutter development in Cursor, Copilot, and other IDEs. Make the AI assistant write clean Flutter code.
★ 670KTX — Context for AI Agents
Allows AI agents (Claude Code, Codex, etc.) to precisely request data via MCP with skills, memory, and a semantic layer. 🔥
★ 668 · TypeScriptACCELERATING CODE REFACTORING
Tura reduces the number of code rewrite steps by 83.1% and increases DeepSWE pass rate by 16.7 percentage points compared to Codex CLI.
★ 667 · RustGET-IT: READ AND UNDERSTAND
The get-it project helps you quickly read, visualize, and extract information from any content. Created at GDG AI Hack Milan 2026.
★ 666 · JavaScriptpr-af - AI Code Reviewer Go
pr-af — open‑source AI code reviewer, #1 in Code‑Review‑Bench. Written in Go, ready to boost PR quality.
★ 666 · GoAI-DIAL-CORE: API FOR ALL MODELS
Unified API gateway for working with various LLMs, assistants, and applications. Java, 658 stars.
★ 663 · JavaRELEVANT PAPERS ON REASONING LLM
Collection of the latest papers and benchmarks on reasoning with foundation models. Enables quick discovery of cutting-edge research and comparison of their performance.
★ 663FLUTTER EDGE AI: RIGHT ON DEVICE
On-device AI for Flutter: Gemma and other LLMs, multimodality, function calling, embeddings, RAG and speech. Supported on Android, iOS, Web, macOS, Windows and Linux.
★ 663 · JavaScriptMULTI-LLM REVEALS CELLS
The project uses coordinated responses of multiple large language models for accurate cell-type annotation in single-cell RNA-seq.
★ 662 · PythonAI ASSISTANT FOR MEETINGS AND INTERVIEWS
Open-source AI assistant for meetings and interviews: recognizes speech on-device, gives instant answers and stays hidden from screen sharing. Available on Windows, macOS
★ 661 · TypeScriptAI AGENT FOR CODING IN RUST
Open-source AI agent for coding and runtime: single binary, any model, native MCP support. Runs in terminal, CI, and as a daemon.
★ 661 · RustAI CHAT IN RESUME
An interactive resume built with React 19 and TypeScript featuring a chat to talk with an AI avatar about my experience.
★ 660 · HTMLGEMINI WEB API – UNLOCK AI
gemini-web-to-api — a reversed API of the Gemini web app, compatible with OpenAI, Gemini, and Claude keys. Enables using AI as a regular API.
★ 660 · GoAGENT SKILL: HUMAN-LIKE TEXT REWRITING
Agent skill that removes AI generation markers and rewrites text per ASD-STE100 Simplified Technical English. Works in two passes: removes 26 AI tells, then shortens sent
★ 659 · PythonAGENT FOR MACOS 26 AI
Agent — AI assistant for macOS 26, controlling the desktop via local and cloud LLMs. Enables task automation, writing scripts and code, interacting with applications. Sup
★ 656 · SwiftGPT-HOME SMART HOME ON RASPBERRY
Local chatbot based on ChatGPT, running on Raspberry Pi via LiteLLM and LangGraph. Your smart assistant without clouds.
★ 655 · PythonMCP Adapter for Pi Agent
A token‑efficient Model Context Protocol adapter for a coding agent on Raspberry Pi that significantly reduces token consumption when interacting with AI.
★ 653 · TypeScriptMIXING RECURSIONS FOR LLM
The Mixture-of-Recursions project offers dynamic recursion depth selection per token in a transformer, enabling adaptive compute allocation. This reduces KV‑cache costs a
★ 651 · PythonPROXMOXMCP-PLUS - MANAGE PROXMOX WITH AI
The project uses MCP and OpenAPI for secure management of VMs, LXCs, backups, and snapshots of Proxmox VE via LLMs and AI agents.
★ 650 · PythonCASBIN-GATEWAY - AI API PROTECTION
Casbin Gateway is an AI- and MCP-oriented security gateway for HTTP traffic with an online demo. Protect APIs in just a couple of clicks.
★ 650 · GoBrowser LLM Device Pool
Run large open‑source LLMs in the browser by linking multiple devices into a peer‑to‑peer compute pool, enabling chat and coding agents without any server.
★ 649 · JavaScriptSHOW-ME-THE-STORY: AI NOVEL
Self-hosted novel generator in Go: a single binary + web UI, compatible with the OpenAI API. It writes chapter by chapter with review, foreshadowing, and fact‑check.
★ 648 · GoPASTEGUARD - YOUR SECRETS SAFE
Open-source proxy for LLM that gives AI context, not your secrets.
★ 646 · TypeScriptDOUBLE CLOUD CODE
headroom-desktop written in Rust lets you double the usage limits of Claude Code and Codex. More agents, fewer restrictions.
★ 646 · RustSmart Screen Streaming to AI
Capture and stream screen content to AI efficiently by filtering unchanged frames and combining multiple frames into single images, minimizing token usage.
★ 644 · JavaScriptMCP-REDIS: AI FOR REDIS
The official MCP server for Redis provides a natural language interface, enabling agents to manage and query Redis data without code.
★ 642 · PythonPRIVATE CHAT AI
Local AI chat in the browser using WebGPU and WebLLM, with no servers or cloud data.
★ 642 · TypeScriptPAPERFORGE — AI SCIENCE IN ONE
PaperForge automates the entire scientific paper writing process: idea generation, literature search, experiments, and LaTeX compilation.
★ 641 · PythonATRIA-DAWN: LLM AGENT OF NEW GENERATION
Atria-Dawn is a framework for autonomous LLM agents with memory, planning and tool invocation. It lets you quickly assemble workflows without boilerplate code. Supports a
★ 640VENDO - AGENTS FOR BUSINESS
The vendo platform provides embeddable AI agents that automate tasks, create views, and connect client tools.
★ 639 · TypeScriptRepo for LLM Unlearning
A collection of papers, code, and resources on machine 'unlearning' in large language models. Helps researchers remove knowledge from LLMs.
★ 638OS PLATFORM AI AGENTS
os — framework for creating AI agents: LLM support, real-time chat and voice, document training, analytics, user management, SSO and widgets. Built with React 19, Redux T
★ 638 · TypeScriptTRUSTLLM: EVALUATION OF LLM TRUSTWORTHINESS
TrustLLM — a toolkit for assessing LLM trustworthiness, presented at ICML 2024. It includes benchmark suites, metrics, and datasets for measuring hallucinations, bias, an
★ 636 · PythonLLM-BASED BANK AUTOMATION
AI automation of bank statements and personal financial analysis with predictions. 583 ⭐ on GitHub.
★ 634 · Jupyter NotebookOLLITERT LLM SERVER ANDROID
OlliteRT turns your Android smartphone into a local inference server compatible with the OpenAI API. It works completely offline, ensuring privacy and open-source code in
★ 633 · KotlinOPTIMUM-INTEL - ACCELERATE INFERENCE WITH INTEL
Optimum Intel provides Intel tools to accelerate AI model inference, simplifying deployment and boosting performance.
★ 630 · Jupyter NotebookBEST SELF-IMPROVING AGENTS LIST
A curated collection of resources on agent self-improvement based on foundation models. Ideal for AI researchers and developers.
★ 630 · TeXCREATE YOUR AI TWIN
agent-me lets you turn your knowledge, memories, and decisions into an open, inspectable AI‑agent twin. The project is written in Python and uses a knowledge graph for st
★ 630 · PythonNANOLANG – Language for LLM Coders
Minimal experimental language in C, created specifically for LLM code generation.
★ 629 · CLLM DIRECTLY IN TERMINAL
Fast stateless LLM for the shell. The qq command answers questions, qa executes commands, written in lightning‑fast Rust.
★ 629 · RustTalos worker client
GPU worker client for the Talos network. It connects to your account, receives open‑model inference tasks via WebSocket, and reports uptime for reward calculation.
★ 629 · PythonRKLLAMA — LOCAL AI ON ROCKCHIP
Ollama alternative for Rockchip NPU. Run AI and Deep Learning models directly on Rockchip devices with NPU optimization.
★ 627 · PythonMEMORIA — MEMORY FOR AI AGENTS
Safe memory management for AI agents on Rust. Protection of data integrity, fewer hallucinations, stable long-term context.
★ 627 · RustVIDEO RECAP WITH CLAUDE CODE
The project provides Claude Code skills for creating Chinese narration recaps from video files. It exports the result as an editable JianYing/CapCut draft. It runs locall
★ 626 · PythonPRISM-INSIGHT - AI STOCK TRADING
AI-based stock analysis system that generates buy/sell signals in real time.
★ 623 · PythonAI PARTNER WITH TERMINAL FORWARD CODING
agentty — a static binary AI pair‑programming assistant for the terminal. It starts in sub‑millisecond and works with any LLM, including Claude via API.
★ 623 · C++CHATS - UNIVERSAL AI GATEWAY
A powerful frontend gateway for LLM with support for 21+ models and simple integration.
★ 621 · C#WAYLAND AI AGENT FOR DESKTOP
Wayland — AI agent on TypeScript, which through Electron perceives the environment, reasons with Claude Code and performs actions in real time. The project uses ACP to or
★ 618 · TypeScriptNew projects on this topic in Telegram
Telegram →