An asynchronous FastAPI-based RAG API that uses pgvector or MongoDB to provide ID-based document indexing and retrieval with built-in ownership security.

xLSTM is a novel recurrent neural‑network architecture that extends LSTM with exponential gating and a matrix memory, enabling large‑scale language models (e.g., a 7 B‑parameter model). The repo provides PyTorch code, custom CUDA/Triton kernels (via the `mlstm_kernels` package), a single‑file 7B implementation, configuration utilities, and pretrained weights on Hugging Face. Installation is via conda + pip, and the library works on NVIDIA GPUs (fast Triton kernels) with fallbacks for AMD or Apple Silicon.

A lightweight Python library for efficient automated machine learning (AutoML) and hyperparameter tuning to find high-quality models with minimal computational resources.

A macOS utility that monitors Claude AI usage limits in real time via the menu bar, desktop widgets, and a floating overlay for Claude Code sessions.

HappyClaw is a self‑hosted platform that runs Anthropic Claude Code agents as long‑living, multi‑user services. It wraps the Claude Agent SDK (TypeScript) and Claude Code CLI, offering web‑based and eight instant‑messaging channel interfaces, isolated workspaces (host or Docker), full Claude Code capabilities (file access, shell, browser, plugins), scheduling, usage tracking, and enterprise‑grade RBAC and backup.

An adversarial AI-agent skill for code review and security auditing that uses a Hunter-Skeptic-Referee pipeline to eliminate false positives.

A benchmark and baseline agent for assessing the ability of LLMs to perform root cause analysis by analyzing telemetry data like logs, metrics, and traces.

An AI assistant for reMarkable tablets that captures handwritten input via screenshots and responds by drawing or typing directly back onto the screen.

A series of ultra-efficient reasoning models (1.5B and 3B parameters) that achieve frontier-level performance in math and coding through a Spectrum-to-Signal post-training methodology.

A photogrammetric computer vision framework that provides algorithms for 3D reconstruction and camera tracking from unordered photographs or videos.

An AI workstation that enables autonomous AI agents to operate a local browser, terminal, and knowledge base to automate complex workflows 24/7.

A lightweight multimodal LLM inference engine optimized for mobile and edge devices, supporting various NPUs and GPUs with advanced quantization.

CasADi is an open‑source library (available on GitHub and PyPI) with extensive download activity. The project’s homepage and install guide provide documentation and installation steps, and you can add it to a Python environment with `pip install casadi`.

A comprehensive toolkit from Hugging Face for evaluating LLMs across multiple backends with support for over 1,000 benchmarks and custom metrics.

A physics-rich simulation environment for robotic vision and interaction, featuring a large-scale dataset of articulated objects for part-level understanding.

A bridge that connects OpenCode to Cursor subscription models, translating prompts and tool calls between OpenCode and cursor-agent.

An MCP server that enables AI agents like Claude to autonomously analyze and trade on Polymarket prediction markets with built-in safety limits.

A Go library for running and fine-tuning ONNX transformer pipelines, enabling Hugging Face compatible inference and training natively in Golang.

A security skills framework and ruleset that embeds secure-by-default practices into AI coding agents to prevent vulnerabilities during code generation and review.

A general-purpose speech recognition toolkit that wraps Whisper models into a REST API for easy deployment and integration.

A comprehensive AI video prompt engineering kit featuring 543 tested prompts, 15 supported models, and automated monitoring of official endpoints to ensure prompting techniques stay current.

MOVA is an open‑source model that generates synchronized video + audio from a text prompt (optionally with a reference image). It uses a dual‑tower video‑audio architecture, provides 360p/720p checkpoints, inference scripts, LoRA fine‑tuning pipelines, evaluation code, and integrations with SGLang, ComfyUI, and a hosted API.

A local-first desktop companion powered by MiniCPM that provides a chat interface and visual reactions to the activity of coding agents.

An ultrafast, Numpy-powered Python implementation of BM25 for high-performance lexical search and document retrieval.

A provider-agnostic personal AI agent that runs in the terminal, providing LLMs with shell, Python, and web access to assist in coding and autonomous knowledge work.

tw‑legal‑rag is an open‑source Python CLI that connects to a hosted semantic‑search service containing 22 million+ Taiwanese judgments, statutes and administrative rulings. It lets you search with natural language, fetch exact case numbers, bundle results (including excerpts, case‑history and citation whitelists) into a JSON file, and run a deterministic citation‑check against LLM‑generated answers. The tool itself never calls an LLM and works without API keys, serving as a retrieval‑augmented generation (RAG) front‑end for legal AI applications.

The ESPHome source code for the Home Assistant Voice: Preview Edition, providing the firmware needed for voice-controlled home automation hardware.

A large-scale similar image search tool that uses feature point matching to find full images using a small cropped screenshot.

A desktop application for Douyin that combines watermark-free video downloading with AI-powered content analysis to automatically tag and summarize videos.

A Model Context Protocol (MCP) server that enables AI applications to securely explore and query MySQL databases through a structured interface.

Jupyter MCP Server is an open‑source bridge that lets AI agents talk to Jupyter notebooks in real time via the Model Context Protocol. It exposes notebook files, cells, and kernel execution as structured tools, supports many cloud sandbox back‑ends, and uses OAuth 2.1 for secure token‑based access.

A tool that automatically generates and maintains AI agent context files to prevent hallucinations and keep AI coding assistants in sync with codebase changes.

A digital data-generation pipeline that synthesizes 4D human-object interaction trajectories from 3D assets and video priors to train humanoid robots in loco-manipulation.

An open standard for shared agent learning that allows AI agents to store and retrieve structured knowledge to avoid repeating mistakes.

A fast, offline OCR library for Node.js and C++ that provides local text recognition for images and PDFs with built-in hardware acceleration.

A high-performance, scalable LLM library written in JAX and Python for pre-training and post-training large models on Google Cloud TPUs and GPUs.

XNNPACK is a highly optimized library of low-level performance primitives used to accelerate neural network inference across ARM, x86, WebAssembly, and RISC-V platforms.

The official PHP SDK for the Model Context Protocol (MCP), enabling developers to build MCP servers and clients to expose PHP tools and resources to AI models.

A collection of specialized React Native knowledge bundles that can be installed into AI coding assistants to help them optimize, test, and migrate apps.

A reference implementation of the agentic registration protocol, enabling AI agents to authenticate with services on behalf of users via identity assertions and claim ceremonies.

An open source framework for the agile development of Chisel-based RISC-V systems-on-chip, integrating processor cores, accelerators, and VLSI tooling.

A self-hosted RSS reader and personal knowledge management tool that uses intelligent aggregation and vector databases to help users manage high-volume reading.

A collection of Claude Code skills designed to automate academic research workflows, including exhaustive literature retrieval, verified proposal writing, and high-fidelity slide generation.

A federated, domain-anchored specification for cataloging and discovering agentic resources like MCP servers, skills, and APIs across networks.

A JavaScript library that enables running pretrained machine learning models directly in the browser using ONNX Runtime, removing the need for a server.

DeepLabCut is a markerless pose estimation toolbox that allows researchers to track animals and objects in video without physical markers using deep learning.

A high-fidelity robot simulation platform combining Unreal Engine 5 and MuJoCo to provide realistic rendering, high-rate physics, and ROS 2-compatible data streaming.

A set of ready‑to‑use “agent skills” that expose ElevenLabs’ speech, voice‑changing, dubbing, and music generation APIs. Install with `npx skills add elevenlabs/skills`, configure an API key, and then any compatible AI coding assistant can call the skills (e.g., text‑to‑speech, speech‑to‑text, real‑time voice chat, voice‑changer, dubbing). The repo provides Python, JS/TS, and CLI SDKs, plus a test suite for trigger and functional validation. Licensed under MIT.