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

A computational pathology toolbox based on PyTorch that provides an end-to-end API for analyzing tissue images, from data loading to visualization.

EdgeYOLO is an anchor-free object detector optimized for real-time performance on embedded edge devices like Nvidia Jetson, supporting multiple deployment formats including TensorRT and RKNN.

Diffuman4D is a spatio-temporal diffusion model for high-fidelity 4D consistent human view synthesis, enabling free-viewpoint rendering from sparse-view videos.

A Python toolkit for creating high-fidelity, privacy-safe synthetic tabular and language data to enable secure data sharing and analysis.

A public leaderboard that uses Vectara's Hallucination Evaluation Model (HHEM) to measure and compare how often different LLMs hallucinate when summarizing documents.

FlowEdit is an open‑source PyTorch implementation of an ICCV 2025 paper that enables fast, inversion‑free text‑guided image editing by reusing pre‑trained diffusion models (e.g., Stable Diffusion 3, Flux) together with a flow network. Users provide a source image, a source caption, and a target caption; the system computes a concise edit code and produces the edited image without costly latent‑space inversion. The repo includes ready‑to‑run scripts, YAML‑based configuration, and links to ComfyUI nodes for broader integration.

A PyTorch implementation of Mean Flows and Improved Mean Flows for high-quality, one-step image generation.

A production-grade deployment toolkit for LLMs and VLMs based on PaddlePaddle, offering high-performance inference and broad hardware compatibility.

A Python-based service for black-box optimization and hyperparameter tuning at scale, providing a distributed client-server infrastructure for finding optimal parameters.

A Python toolbox for machine learning on partially-observed time series, providing unified APIs for imputation, forecasting, and anomaly detection on data with missing values.

An extensible open-source library for Python and R that provides metrics and algorithms to detect and mitigate algorithmic bias in machine learning models.

An open-source library built on Apache Spark that provides scalable, distributed APIs for machine learning tasks including text analytics, vision, and anomaly detection.

A fast, flexible C++ machine learning library providing a core tensor interface and neural network tools for high-performance AI research across speech, vision, and text.

Swiftide is a Rust framework for building LLM applications with three core components: an agent harness for semi-autonomous agents, typed task graphs for workflow orchestration, and streaming RAG pipelines for indexing and retrieval.

AGiXT is an AI automation platform that uses natural language to control digital and physical environments through a wide array of built-in extensions and multi-provider support.

An open-source LLMOps platform for prompt management, observability, and caching to optimize AI operations and reduce costs.

An intelligent document automation system that uses YOLO, OCR, and LLMs to extract and analyze financial data from bank statements with a focus on local LLM execution and PII privacy.

A native iOS and macOS workbench for testing and shipping with Apple's Foundation Models framework, featuring a prompt playground, tool integration, and adapter comparison.

A set of helper libraries for NestJS that provides high-performance validation, automatic SDK generation, and AI-powered development tools for typed API servers.

A federated AI search and RAG platform that queries data sources live, eliminating the need for vector databases and ETL pipelines while keeping data in place.

A unified, tileable workspace that integrates AI agents, code editors, web browsing, and document editing to eliminate context switching for complex workflows.

An open-source AI orchestration framework that implements an agentic agile workflow to prevent context loss and planning inconsistency in AI-assisted software development.

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

A desktop interface for OpenClaw AI agents that replaces command-line orchestration with a graphical user interface for managing agents, channels, and automated tasks.

A decentralized P2P network where autonomous AI agents collaboratively conduct ML research, pool compute for distributed training and inference, and share discoveries via a global leaderboard.

HMMER is a biological sequence analysis tool that uses profile hidden Markov models to search databases for homologous sequences.

An application for removing logos and watermarks from videos, featuring automatic detection and the ability to handle moving logos.

A Python framework for multi-objective optimization that provides a wide range of evolutionary algorithms, benchmark problems, and tools for statistical analysis.

A machine learning library that rapidly benchmarks dozens of classification, regression, and time series forecasting models to identify the best performing algorithm without manual tuning.

An end-to-end autonomous AI research system that autonomously conducts literature reviews, generates hypotheses, and performs experiments to produce SOTA scientific results.

Holistic Trace Analysis (HTA) is a Python library that reads PyTorch Profiler (Kineto) traces from distributed training jobs and produces data‑frames and visualisations showing where GPUs spend time—compute, communication, memory, or idle. It offers kernel‑level breakdowns, idle‑time causes, comm‑compute overlap, frequent kernel patterns, launch‑time stats, augmented memory‑bandwidth/queue‑length counters, and a trace‑comparison tool. Install via `pip install HolisticTraceAnalysis` (Linux/macOS, Python ≥ 3.10) and use the `TraceAnalysis` class in a notebook to get pandas DataFrames for each analysis. Documentation, examples, and an experimental CUPTI counter API are provided.

A scikit-learn compatible library of Relief-based feature selection algorithms that identify important feature interactions in tabular data without exhaustive searching.

A visual node editor for creating and prototyping OpenCV algorithms with live pipeline previews.

An AI-agent plugin that automates Japanese income and consumption tax filing, from bookkeeping and tax calculation to e-Tax form submission, for individuals and small business owners. It works with Claude Code, Cursor, and 40+ other AI coding agents.

autoresearch‑claude‑code is a Python framework that runs budget‑constrained auto‑research loops with Claude Code or Codex agents. You define a goal, a deterministic scorer, and files the agent may edit; the system creates a Git branch, measures each change, and automatically commits only statistically‑significant improvements while respecting run‑time and turn limits.

An open-source document intelligence API that converts PDFs, PPTs, Word docs, and images into structured, RAG-ready chunks using layout analysis and OCR.

A fast hybrid vision transformer that uses structural reparameterization to achieve low-latency image classification on mobile devices.

A deep-learning tool for extracting instrumental tracks from songs by separating vocals from the background music.

An agentic AI system for cybersecurity research and defense that uses multi-agent orchestration, MCP, and ACP to analyze security reports and source code.

DeepProbLog is a Python library that merges ProbLog’s probabilistic logic programming with deep learning by introducing neural predicates—facts whose probabilities are output by PyTorch models. It supports exact and optional approximate inference, includes example experiments from its research papers, and can be installed via `pip install deepproblog`.

An enterprise-grade biomedical research tool that automates the search and synthesis of data from PubMed, FDA labels, and patents into cited research reports.

A family of specialized AI coworkers for non-developers to handle Korean business tasks, featuring a PM agent that automates project setup and skill chaining via natural language.

NeuMan is a neural radiance field implementation that reconstructs an animatable human and their background scene from a single video, enabling novel view and pose synthesis.

A 3D virtual assistant SDK for Unity that turns 3D VRM models into voice-enabled chatbots with synchronized speech, lip-sync, and autonomous animations.

A set of C# libraries and tools that enable .NET applications and Unity projects to communicate with ROS nodes for robotics simulations and automation.

A real-time multi-object tracking system written in C++ that implements ByteTrack, OC-SORT, Deep OC-SORT, and DeepSORT using YOLO and ONNX Runtime.

MobileOne provides PyTorch implementations of five ultra‑fast CNN backbones (S0‑S4) that achieve 71‑79 % ImageNet Top‑1 accuracy with ≤ 2 ms latency on an iPhone 12 Pro. The repo includes training‑ready and inference‑ready checkpoints, CoreML models for on‑device deployment, and an iOS benchmark app (ModelBench). Re‑parameterization fuses training‑time branches into a single fast network for inference.