A local-first microVM sandbox that provides AI agents with a secure, disposable Linux environment to execute code and install tools without affecting the host system.

A Transformer-based model for encrypted network traffic classification that learns datagram contextual relationships through pre-training and fine-tuning.

A Python library that transcribes audio using faster-whisper and uses LLMs to translate and polish the text into synchronized .lrc subtitles.

A provider for the Vercel AI SDK that enables seamless access to over 300 LLMs and embedding models via the OpenRouter API.

A complete Rust SDK for building high-performance coding agents with built-in tool execution, graph memory, and sub-agent orchestration.

A Claude Code skill that applies a professional advertising methodology to generate witty, minimalist marketing copy and posters using a split AI-image and code-based layout pipeline.

A ROS 2-based error-state Extended Kalman Filter (EKF) for high-precision localization that fuses GNSS, IMU, and odometry to estimate position, velocity, and attitude.

A ROS 2 integration for nvblox that enables real-time 3D scene reconstruction and 2D costmap generation for vision-based obstacle avoidance in robots.

A framework for controlling robots within the ROS 2 ecosystem, providing a standardized way to manage robot hardware and control loops.

MCAP is a serialization-agnostic logging library and file format designed for efficient data recording and analysis in robotics applications.

A set of libraries and tools that enable developers to write ROS 2 applications in Rust, providing a high-performance and memory-safe alternative to C++ and Python.

A document conversion tool that turns PDFs, Office files, and images into AI-ready Markdown using multimodal vision models, supporting local inference for privacy and cost-efficiency.

A series of video foundation models and large-scale datasets designed for multimodal video understanding, long-context modeling, and contextual reasoning.

An MCP server that enables structured, spec-driven development for AI agents by enforcing a sequential workflow of requirements, design, and tasks with a real-time monitoring dashboard.

An AI-powered research tool that automates the gathering of information from the web and local files to generate comprehensive, in-depth reports.

A vision-language RAG pipeline that embeds documents as images to prevent information loss from text parsing and uses evidence-guided reasoning for multi-image QA.

Astrid is an operating-system project for WASM-based capsules that provides an OS-grade security boundary for AI agents, separating decision-making from execution authority.

An AI fitness assistant that uses RTMPose for real-time pose detection and automatic exercise repetition counting via a webcam.

VecturaKit is a Swift library that provides an on‑device vector database with hybrid vector + BM25 search. It supports multiple embedding back‑ends (Apple NaturalLanguage, OpenAI‑compatible APIs, swift‑embeddings models, and Apple MLX), custom storage/search plug‑ins, and CLI tools, enabling privacy‑first semantic search in iOS/macOS/watchOS/tvOS/visionOS apps.

An open-source backend for MongoDB that instantly generates REST, GraphQL, and WebSocket APIs with built-in authentication and no backend code required.

A multi-turn multimodal RAG framework that uses agentic reinforcement learning to enable VLMs to reason and retrieve information across text, images, and videos.

A unified multimodal backend for AI data apps that combines a database, orchestration, and serving layers into a single Python file.

A collection of reference architectures and examples for deploying and operating distributed AI training and inference on AWS infrastructure.

An extension of vLLM that enables hybrid GPU and NUMA-aware CPU inference for MoE models, allowing them to span across VRAM and system memory.

A modern C++ library for high-performance GPU inference of CNN models using NVIDIA TensorRT, featuring a safe engine cache and zero-copy Python bindings.

Oumi is an open‑source Python platform that covers the entire lifecycle of foundation models—data prep, training (SFT, LoRA, QLoRA, GRPO), evaluation, and deployment—via a unified CLI and a library of ready‑made YAML recipes for many text and vision‑language models.

A neuro-symbolic framework for building and optimizing LM applications like RAGs and agents using a Keras-like declarative API and in-context reinforcement learning.

A fully localized Chinese distribution of OpenClaw, providing a translated CLI and Dashboard for managing a personal AI assistant platform.

A low-code development tool for building multi-agent LLM applications that simplifies the process of prototyping, deploying, and iteratively optimizing AI workflows.

A PyTorch-based toolbox for Image Quality Assessment (IQA) that provides GPU-accelerated implementations of numerous full-reference and no-reference quality metrics.

A local-first communication and coordination layer that allows different AI coding agents to synchronize work, claim files, and exchange messages within a shared repository.

A token-efficiency skill for coding agents that reduces API costs and token usage by teaching agents more efficient lookup and polling habits.

A macOS desktop app for stock market analysis that combines local technical indicators with verifiable AI insights and a SEPA strategy dashboard.

A unified runtime and workbench for AI agents that provides Desktop, Web, and TUI interfaces for managing local development tools and plugins.

A general-purpose foundation GUI agent that unifies mobile, web, and desktop interaction through a closed-loop perception-reasoning-action system.

Face Anything is a feed‑forward neural model that reconstructs a temporally‑consistent 4‑D (3‑D + time) face from any image sequence or single photo, outputting point‑cloud videos, depth/normal maps, canonical coordinate visualisations, per‑frame PLY meshes and camera data. It runs on a CUDA GPU, installs via a one‑line script, and offers configurable processing modes and background removal.

A bridge that connects WeChat and Enterprise WeChat to local AI coding CLIs like Claude Code and Codex, enabling remote interaction and approval of local terminal sessions.

An AI agent system that automates the end-to-end mathematical modeling process, from problem analysis and model selection to code execution and academic report generation.

A Python‑based Agent Skill that lets LLM agents interact with Kaggle: fetch competition info, summarize top write‑ups, search discussions, download and reproduce notebooks, submit solutions, and upload datasets—all via natural‑language commands.

An AI-powered video creation platform that automates the workflow from script generation and character design to storyboard creation and final video rendering.

Rivet Dynamic Apps is a framework for securely executing user‑provided request‑handler code in sandboxed V8 isolates, with versioned state stored in Rivet actors and a fast cache‑hit path for low‑latency responses. It targets AI/LLM workloads by allowing custom, updatable logic to run safely at the edge.

The source code for the firmware used to operate and operate the Crazyflie range of nano-quadcopters, including the Crazyflie 2.x and Roadrunner.

A local web-development stack manager for Windows and macOS that provides one-click services and AI-driven control via an MCP server.

A fast, scriptable CLI for macOS Reminders designed for power users and AI agents, providing deep access to task metadata and private API integrations.

A lightweight, high-performance WebAssembly runtime that provides a secure sandbox for running LLMs, microservices, and plug-ins across edge devices and servers.

An unofficial repackaging project that brings the official Claude Desktop app to a wider range of Linux distributions via .rpm, AppImage, and Nix flakes, adding Linux-specific fixes and diagnostics.

An open-source machine vision platform that enables beginners to implement AI and image processing on microcontrollers using Python3.

Kite is an open-source Kubernetes workspace for multi-cluster operations, observability, and AI-assisted troubleshooting.