An analytics engine that analyzes billions of GitHub events to track trends, rankings, and developer activity in the AI-native open source ecosystem.

A deep learning-based library for identifying the gap coordinates and confidence levels in slider CAPTCHAs.

A collection of reference GPU kernels and problem specifications used to evaluate and benchmark custom kernel implementations in GPU Mode competitions.

A self-hosted web API that provides free, unlimited access to various LLM providers through a single HTTP interface, eliminating the need for paid API keys.

TianRuo OCR v6 is a Windows desktop app that captures screen text, runs OCR (online or offline via PaddleOCR/RapidOCR), and translates the result using many online services. It offers normal, silent, screenshot and clipboard‑listen modes, configurable shortcuts, and supports printed, table and handwritten text.

A machine learning runtime written in Rust that enables running ONNX models trained in PyTorch and other Python frameworks in Rust and WebAssembly environments.

A modular platform for managing uncrewed vehicles and robots from an onboard computer, supporting various autopilots and hardware architectures.

An open testing platform for Kubernetes that allows engineering teams to run, orchestrate, and analyze any automated tests directly within their cluster infrastructure.

A neural network library for JAX that enables object-oriented model definition while preserving JAX's pure function transformations.

A visual data-flow environment for battle-testing and evaluating LLM prompts across multiple models and parameter permutations.

A multi-LoRA inference server that allows users to serve thousands of fine-tuned LLMs on a single GPU to reduce serving costs.

A high-performance ray tracing library by Intel that provides optimized kernels for CPUs and GPUs to accelerate photo-realistic rendering applications.

A Sim-to-Real locomotion starter kit for the Unitree Go2 quadruped robot that adapts the walk-these-ways framework to the unitree_sdk2 architecture.

DART (Dynamic Animation and Robotics Toolkit) is an open‑source physics engine for robotics, animation, and ML‑driven simulation. It offers research‑grade dynamics via Featherstone’s algorithm, supports URDF/SDF/MJCF model loading, and provides C++ and Python APIs. Stable DART 6 is production‑ready; DART 7 is an experimental redesign. Install via Conda, PyPI, or source builds with `pixi`.

A centralized guardrail system for AI agents that enforces runtime safety rules to block prompt injections and PII leaks without changing agent code.

A collection of ready-to-use public datasets provided as tf.data.Datasets to simplify the process of loading and preparing data for machine learning.

A minimal and safe Rust API wrapper for the NVIDIA CUDA toolkit, providing safe abstractions for GPU acceleration and kernel management.

AWorld is an agent orchestration framework that allows experts to encode their domain knowledge into reusable skills to build and evolve autonomous, high-precision AI agents.

An open-source platform for generating secure full-stack apps and AI agents from natural language, featuring a compiled runtime and built-in MCP server for self-hosted deployment.

A user interface for LangGraph agents that allows humans to review, edit, and respond to agent interrupts in a centralized inbox.

A self-hosted security camera platform providing live viewing, recording, and on-device AI detection while keeping footage on user-owned hardware.

An Android app that enables fully offline text chat and Stable Diffusion image generation using local runtimes like ONNX, MediaPipe, and MNN.

A generative model that samples the equilibrium distribution of protein monomer structures from amino acid sequences, featuring a steering system to ensure physical plausibility.

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

A Python library for defining and optimizing symbolic mathematical expressions with multi-dimensional arrays, serving as the computational backend for PyMC.

Official ROS 2 packages for the ROBOTIS OpenMANIPULATOR platform, providing the tools needed to control the robot arm and develop physical AI applications.

A multi-vector search engine (NextPlaid) and a corresponding semantic code search tool (ColGREP) that provide high-precision retrieval by storing multiple embeddings per document.

A Transformer-based SLAM system that unifies localization and mapping into a single model for 3D reconstruction and camera tracking from image sequences.

An image super-resolution and restoration tool specifically optimized for ACGN illustrations, CG, and manga using deep learning models.

An open-source, self-hosted AI agent runtime built with C#/.NET that can be embedded into applications or used as a desktop app to manage extensible AI agents.

An automation tool and MCP server for Google NotebookLM that enables creating notebooks, performing grounded research, and generating artifacts via CLI or AI agents.

WGAI is a Spring‑Boot + Vue web platform that bundles vision (OpenCV, YOLO, OCR, face ID), audio (speech‑to‑text, ChatGPT‑style dialogue), digital‑human avatars, and AGV navigation into an offline‑first, industrial‑grade AI management system. It offers online data annotation, model training, real‑time video/audio analysis, and system monitoring, with a UI for configuration and a public demo instance. The code is AGPL‑3.0 and support is mainly via a paid Chinese community.

An auditable agent-team workflow for coding agents that turns a single AI into a structured engineering team with event-sourced runtime and quality gates.

A PyTorch domain library for fine-tuning and using pretrained Geospatial Foundation Models for tasks like image segmentation and classification.

An API-first platform for enterprise document intelligence that converts invoices, receipts, and statements into structured JSON using Vision LLMs and agentic workflows.

A browser extension that uses AI to filter unwanted posts from Twitter/X feeds based on natural language descriptions of topics to avoid.

A Python library for Difference-in-Differences causal inference providing a wide range of econometric estimators for measuring campaign lift, policy impact, and staggered rollouts.

A collection of specialized knowledge modules (skills) for AI coding assistants to improve their proficiency in Vue, Nuxt, and the NuxtHub ecosystem.

A self-hosted development environment for building and orchestrating modular, production-grade data pipelines using a notebook-style interface.

A workflow orchestration system for Pi that enables the execution of repeatable, multi-step AI agent workflows for tasks like research and code review.

Datalevin is a fast, durable Datalog database that provides a declarative alternative to SQL and graph databases, featuring built-in vector search and AI-native capabilities.

A production-focused Java library for extracting structured tables from text-based, scanned, and image-heavy PDFs using OCR and table structure detection.

A synthetic micro-benchmark tool for measuring the peak compute performance (TFLOPS/TOPS) of CPUs, GPUs, and NPUs across multiple backends like CUDA, ROCm, and ONNX.

A PyTorch collection of deep learning models for remote sensing semantic segmentation, featuring single-modal, multimodal fusion, and unsupervised domain adaptation techniques.

A PyTorch-based library for 3D Computer Vision research providing differentiable rendering and efficient tools for manipulating 3D meshes and point clouds.

A high-performance computing platform for systems analysis and multidisciplinary optimization that enables the decomposition and parallel solving of complex models.

A multi-container Slurm cluster deployed via Docker Compose for rapid development, testing, and lightweight HPC job scheduling.

A framework for analyzing large-scale single-cell perturbation experiments, providing tools to harmonize data and analyze how cells respond to genetic, drug, or environmental stimuli.