A Chinese-language version of the CLIP model trained on 200 million image-text pairs for cross-modal retrieval and zero-shot image classification.

Toolathlon is a benchmark for assessing language agents' general tool-use capabilities across 600+ diverse tools in realistic, long-horizon software environments.

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.

A score-regularized continuous-time consistency model for distilling large video diffusion models, enabling high-quality, diverse video generation in 2-4 steps.

Gaussian-LIC2 is a photo-realistic SLAM system that uses LiDAR-Inertial-Camera fusion and Gaussian Splatting to perform real-time pose estimation and 3D mapping.

A comprehensive resource hub and benchmark for Infrared and Visible Image Fusion (IVIF), providing curated datasets, a categorized catalog of fusion methods, and evaluation tools.

A local-first control plane and dashboard for managing multiple AI coding agents, providing unified session browsing, provider bridging, and task orchestration.

An AI governance platform that helps businesses manage AI compliance, risk, and security through on-premises hosting and support for frameworks like the EU AI Act.

A lightweight C++/Python 3D viewer and UI for quickly visualizing meshes and point clouds with associated scalar and vector data.

A collection of professional software engineering skills for AI coding agents to transform junior-level code into senior-quality software using SOLID, TDD, and clean code practices.

A collection of Docker Compose demos for deploying and orchestrating AI agents, supporting multiple frameworks and local or remote model execution.

An open-source singing synthesis editor for mobile devices that supports DiffSinger, UTAU, and Vogen voicebanks.

An MCP server that enables AI assistants to read, search, organize, and send emails using IMAP and SMTP.

A self-hosted AI coding workspace that orchestrates multiple AI agents from one interface, providing sandboxed environments and multi-agent workflows.

An open-source DIY golf launch monitor that uses Raspberry Pi and machine learning to track ball launch speed, angles, and spin.

A nested U-Net architecture for medical image segmentation that improves upon the original U-Net by redesigning skip connections and allowing for flexible network depth.

A family of multi-modal, spatio-temporal foundation models for Earth Observation, providing a scalable framework for planetary intelligence using satellite data.

OpenSim is a software framework for developing musculoskeletal models and creating dynamic simulations of movement to study human and animal biomechanics.

Dify-Sandbox is a secure execution environment for running untrusted code in multi-tenant systems, restricting resource and system call access to protect the host.

chipStar is a compiler and runtime layer that enables CUDA and HIP applications to run on any platform supporting SPIR-V, such as Intel, ARM, and RISC-V GPUs.

An affordable motor controller with integrated encoder and CAN bus for the precise control of 3-phase brushless motors.

A macOS menu bar app that automatically monitors and aggregates real-time usage and billing for multiple AI providers in one unified dashboard.

PLINDER is a comprehensive, annotated dataset and evaluation resource for training and and benchmarking protein-ligand docking algorithms in structural biology.

An MLIR-based compiler framework for co-designing domain-specific languages and architectures to optimize the deployment of AI models like DeepSeek R1 and Whisper.

A local management interface for AI coding agents that organizes Claude Code and ChatGPT Codex sessions into a unified, scrollable browser-based terminal with cost tracking and workflow tools.

A cognitive memory database that implements biological memory primitives like Hebbian learning and temporal decay to provide AI agents with persistent, evolving memory.

A minimalist Python library for building and teaching self-driving robots, featuring support for deep learning, computer vision, and GPS autopilots.

An ML-based tool for master data management and entity resolution that deduplicates and links records across disparate data sources at scale.

A cross-platform OpenGL-based 3D game engine featuring extensive procedural generation for universes, cities, and terrain, and high-performance rendering of large models.

NetLogo is a programmable modeling environment for creating agent-based simulations to study complex systems and emergent behavior.

An AI-powered tool that transforms plain English descriptions into verified, structured datasets from the live web with automatic refresh schedules.

AI Team OS is an open‑source platform that turns a Claude‑style LLM into an autonomous “CEO”. It continuously runs without prompting, orchestrating specialist agents, a task wall, research pipelines, and a two‑layer memory system. The FastAPI + React dashboard shows live workflows, decision traces, and a unified search over all memos. It’s aimed at technical founders or research teams who want an AI‑first, self‑driving company‑like workflow.

A free and fast computer vision solution tailored for robots in the FIRST Robotics Competition, providing tools for easy integration and vision processing.

A platform for running a fleet of private coding agents on your own infrastructure, using a spec-driven Kanban board to manage and review their work.

A local-first AI desktop app that provides a shared workspace for multiple coding agents to collaborate, plan, and execute tasks without interfering with each other.

Sudachi is a Java‑based Japanese morphological analyzer that tokenises text, tags parts of speech, and normalises spelling. It supports three segmentation modes (short, medium, named‑entity), large UniDic‑based dictionaries (small, core, full), a plugin system for custom processing, and can be used via command line, Maven library, Elasticsearch plugin, or through Python/Rust ports.

An MLX-based implementation of Nvidia's Parakeet ASR models for Apple Silicon, providing efficient local audio transcription with streaming and timestamp support.

A Microsoft‑maintained, MIT‑licensed tutorial series that teaches JavaScript developers to build generative‑AI apps. It covers LLM basics, prompt engineering, structured output, RAG, tool calling, and the Model Context Protocol, with code samples, quizzes, videos, and a companion chat app. You can start instantly in a GitHub Codespace or run the code locally.

A Rust-based terminal UI and LLM harness that allows AI agents to automate tasks in real Chrome, headless Chromium, or the cloud.

A Python library for predictive uncertainty calibration and conformalization, providing rigorous uncertainty estimations for any machine learning model across various tasks.

An open-source framework for training LLM-based multi-agent systems using reinforcement learning, featuring tree search augmentation for complex reasoning and code generation.

A benchmark for evaluating AI on unsolved, open-ended, and diverse computer science research and algorithmic problems to move beyond saturated textbook benchmarks.

BCPD is a MATLAB‑compatible, command‑line suite for Bayesian non‑rigid registration of point clouds and functions (DET), with accelerated “++” modes, multiple kernel options, and demos for 3‑D reconstruction, shape transfer, and spatial‑omics data.

A Windows desktop application built with Tauri and Rust to monitor DeepSeek API account balances, monthly spending, and token usage trends.

A browser-based tool that converts code repositories from GitHub, GitLab, Azure DevOps, or local files into plain text for use in LLM prompts.

A customizable vim-style status line for Claude Code that provides real-time tracking of token usage, costs, git status, and context window saturation.

A world-champion Shogi AI engine that provides a high-performance search and evaluation framework compliant with the USI protocol.

noteDigger is a pure‑JavaScript web app for audio‑to‑MIDI transcription. It lets users upload audio/video, visualises a spectrogram (STFT/CQT, GPU‑accelerated), runs on‑browser neural‑network models for pitch and instrument‑separation transcription, and provides extensive manual editing tools with keyboard shortcuts. Users can export quantised MIDI files or save a project for later work, all without external libraries or server‑side processing.