A Go CLI for evaluating AI agent skills by scaffolding evaluation suites, running benchmarks, and comparing results across different models.

A GPU-accelerated elevation mapping framework for legged robots that fuses multi-modal data like geometry and semantics to create accurate real-time terrain maps.

A JAX-native library for Multi-Agent Reinforcement Learning that provides GPU-accelerated environments and baseline algorithms for efficient benchmarking.

An SDK that connects AI agents to Slack, Microsoft Teams, and Discord, enabling native interactive UI and human-in-the-loop approvals.

MolmoAct2 is a family of action reasoning models that connect vision-language backbones to flow-matching action experts for real-world robot control and manipulation.

An AI-powered analysis tool for Android Perfetto traces that uses natural-language queries to identify performance bottlenecks like jank and ANRs with evidence-backed reasoning.

A mobile-friendly, web-based coding agent that supports multiple models and modalities to help developers write and manage code.

A lightweight, Kubernetes-native LLM inference platform that provides modular control and data planes for scalable, cost-efficient model serving.

GaussianGPT is a transformer-based model that generates 3D Gaussian scenes via next-token prediction, supporting controllable 3D scene generation, completion, and outpainting.

An on-device AI inference engine that enables the local and offline execution of GGUF-format LLMs across multiple platforms and languages.

A 23-state Unscented Kalman Filter for outdoor robots that fuses IMU, wheel encoders, GPS, and visual SLAM to provide stable localization with built-in diagnostics to prevent silent drift.

An open-source, local-first desktop workbench for AI video that lets users combine multiple generation sources and local ComfyUI to lower production costs.

A local-first workbench for managing prompts, skills, and AI programming assets, featuring one-click distribution to multiple AI coding tools and built-in version control.

A high-performance A2UI SDK for building native generative UI experiences on iOS, Android, and HarmonyOS, enabling LLMs to render interactive components in real-time.

A multi-model AI workspace for experts that allows users to bring their own API keys to access hundreds of models while providing tools for multi-model validation to reduce hallucinations.

An open-source agentic AI toolkit and Vietnamese NLP library that provides multi-provider LLM support and specialized text processing tools for the Vietnamese language.

A proposal to standardize the use of /llms.txt files to provide AI agents with concise, Markdown-formatted summaries and links to clean content on websites.

A memory-centric AI writing desktop system for long-form novels that prevents plot holes and character drift through structured memory ingestion and hybrid retrieval.

An automated AI video and podcast generator that converts long-form YouTube videos into viral shorts and creates AI-powered podcasts from text prompts.

A collection of tools and APIs for content moderation and security threat intelligence, featuring image and video hashing algorithms for similarity matching.

An efficient probabilistic 3D mapping framework based on octrees that allows robots to represent and track occupied, free, and unknown space in 3D environments.

CrewAI Studio is a Streamlit‑based GUI that lets you design, run and export multi‑agent AI crews built on the CrewAI framework, supporting several LLM providers and custom tools, with easy installation via Conda, virtualenv, Docker or one‑click deployment.

AnyLanguageModel is a Swift package that replaces Apple’s FoundationModels import with a unified API for many LLM providers (Apple, Core ML, MLX, llama.cpp, Ollama, OpenAI, Anthropic, Gemini, etc.). It offers a `LanguageModelSession` for prompting, guided generation of Swift structs via `@Generable/@Guide`, tool‑calling support with delegate hooks, optional image inputs, and trait‑based dependency selection to keep binaries small. Installation is via SwiftPM; enable back‑ends with traits (`CoreML`, `MLX`, `Llama`). Security guidance warns against hard‑coding API keys, recommending either user‑provided keys stored in Keychain or a proxy server. The library targets Swift 6.1+, iOS 17/macOS 14+, and works on‑device as well as with cloud APIs.

A local codebase indexing engine that reduces AI coding assistant costs by replacing full-file reads with targeted semantic search and retrieval.

A durable, versioned workspace provider for disposable AI agent sandboxes that ensures artifact lineage and state survival through atomic metadata control.

An open-source toolkit for general OCR, providing lightweight models for unified text, formula, and table recognition and document parsing.

A utility library for Unsloth that enables faster and more memory-efficient fine-tuning of LLMs, vision, and TTS models with significantly reduced VRAM requirements.

An end-to-end Vision AI application and library for rapidly building, annotating, training, and deploying computer vision models optimized for edge hardware.

A CLI proxy that filters verbose shell output before it reaches AI coding assistants, reducing token usage by 60-90% through declarative YAML pipelines.

An open-source, AI-powered presentation generator that creates customizable slides from a topic via an outline-first workflow.

A unified web interface and manager for a wide variety of open-source text-to-speech, audio generation, and audio conversion AI models.

A Python package that adapts the Segment Anything Model (SAM) for geospatial data, enabling easy segmentation of satellite imagery using text, points, or boxes.

A Python library for financial econometrics that provides tools for volatility modeling (ARCH/GARCH), unit root testing, and cointegration analysis.

An automated video distribution pipeline that handles everything from file monitoring and AI-generated captions to multi-platform uploading using browser automation.

SkillNet is an open‑source infrastructure that treats AI agent capabilities as reusable, searchable software assets. It offers a public skill library, a Python SDK/CLI, and a public REST API for discovering, downloading, creating, evaluating, analyzing, and orchestrating these skills. The platform enables agents to reuse existing functionality (e.g., PDF parsing, web‑scraping) instead of reinventing it, provides quality scores, builds skill‑relationship graphs, and can generate scene‑specific prompts for downstream LLM agents. Installation is as simple as `pip install skillnet-ai`; no API key is needed for search or public downloads, while creation/evaluation/orchestration require an LLM API key. The project includes a web explorer, MCP server, and integrations with Claude, OpenAI, OpenClaw, and JiuwenClaw, plus benchmark scripts for ALFWorld, WebShop, and ScienceWorld.

A multisensor-aided inertial navigation system that flexibly fuses IMU, wheel encoders, cameras, GNSS, and LiDAR for robust robot localization and online sensor calibration.

A lightweight, self-hosted database workspace for exploring, editing, and querying data with optional AI-powered natural language interfaces.

A PyTorch library for Conditional Flow Matching, providing simulation-free training for continuous normalizing flows to enable fast generative modeling of images, tabular data, and biological dynamics.

A lightweight digital human model designed for real-time talking-head generation on mobile devices using personalized training from short videos.

agentUniverse is an Apache‑2.0 Python framework for building and orchestrating LLM‑powered multi‑agent applications. It offers ready‑made collaboration patterns (PEER, DOE), supports dozens of LLM providers, includes tools for domain knowledge injection, observability via OpenTelemetry, and a visual workflow UI. The library is used in real‑world financial products at Ant Group and provides extensive docs, sample apps, and a community on GitHub/Discord.

A real-time LiDAR-based SLAM system designed for high-dynamic motion, providing robust odometry and mapping for legged robots and other mobile platforms.

py‑xiaozhi is a Python‑based, async‑first framework for building multimodal AI assistants that can listen, speak, see, and control hardware. It offers real‑time Opus voice streaming, offline wake‑word detection, vision‑language integration, a JSON‑RPC tool server, and secure WebSocket/MQTT communication. Runs on Windows/macOS/Linux desktops and on edge boards (Raspberry Pi, Jetson Nano). Provides GUI (PySide6/QML), CLI, and GPIO modes, and a plugin system for easy extension.

A multimodal toolbox for evaluating embeddings and retrieval systems, providing a standardized benchmark and leaderboard for comparing model performance.

An open-source voice AI orchestration platform for agencies and enterprises to build and deploy scalable, real-time voice agents without vendor lock-in.

An AI assistant featuring Live2D anime avatars that separates character roleplay from background agent tasks to provide both emotional companionship and technical productivity.

A GitHub Action that integrates Gemini AI into development workflows to automate PR reviews, issue triage, and provide conversational AI assistance directly in GitHub.

Crane is a Rust‑based inference framework (built on candle) that runs LLMs, vision‑language models, OCR, ASR, TTS and music transcription fast on CPU, CUDA, Metal and experimental ROCm. It loads GGUF checkpoints directly, provides a clean Rust API and an OpenAI‑compatible HTTP server, and ships ready‑to‑use demo binaries for chat, image chat, OCR, speech, etc.

A repository for experimental and community-contributed extra modules for OpenCV, serving as a testing ground for features before they move to the official core distribution.