A local vision-based monitoring system that detects 3D print failures in real-time to automatically pause printers and alert users, avoiding wasted filament.

A unified collective communication library designed for high-performance AI/ML and HPC workloads across CPU, GPU, and DPU hardware.

A bridge service that lets you run and control AI programming CLI tools like Claude Code and Gemini CLI remotely via WeChat using the official ClawBot API.

A command-line tool for data versioning on cloud storage that allows users to snapshot and switch between data versions without requiring a separate server.

A living benchmark for symbolic regression that provides a unified framework and reproducible results to compare modern SR methods across diverse datasets.

A certification program that defines standardized capabilities for Kubernetes platforms to ensure AI and machine learning workloads are portable and reliable across different clusters.

Open Simulation Interface (OSI) is a generic, protocol-buffer-based interface for the environmental perception of automated driving functions in virtual simulation scenarios.

An Obsidian plugin that enables text search within images and PDFs by using local OCR engines like Tesseract.

TFB is an open-source library for the fair and comprehensive benchmarking of time series forecasting methods, providing a standardized pipeline to prevent evaluation biases.

A curated marketplace of universal skills for AI coding agents, allowing developers to install a single skill and use it across multiple different agent platforms.

A comparative framework for multimodal recommender systems that simplifies the integration of auxiliary data and the benchmarking of various recommendation algorithms.

A framework for Ruby on Rails that introduces Agent-Oriented Programming, allowing developers to build modular AI agents as controllers to manage LLM interactions and tool usage.

An MCP server that enables AI agents to control macOS applications and system functions by executing AppleScript and JavaScript for Automation (JXA) scripts.

GroundX On‑Prem is an open‑source, Kubernetes‑native deployment of EyeLevel AI’s RAG pipeline (document ingestion, encrypted storage, and vector‑augmented search). It ships Helm charts, optional AWS Terraform scripts, and detailed node‑group requirements so you can run the whole system—including GPU‑accelerated vision models and a re‑ranker—inside a secure, air‑gapped cluster.

A Neovim plugin that provides on-demand LLM text completion, inline editing, and chat sessions while ensuring the user maintains full control over API requests.

An extension for Gemini CLI that uses AI to analyze code changes for security vulnerabilities and scans dependencies for known risks.

A word-level English curved text dataset and evaluation framework for improving the detection of arbitrary-shaped text in natural scene images.

An open-source framework for building AI agents that run in the browser, enabling developers to easily embed intelligence into React and Angular components.

An AI-powered equity research tool for Claude Code that generates institutional-grade investment reports with fundamental and technical analysis.

QuietHarness is a tiny, open‑source safety layer for AI‑coding agents (Claude Code, Codex, Cursor). It provides shared configuration templates, dry‑run installation scripts, automatic backups, and risk‑gated actions so the agent won’t overwrite files, skip tests, or perform irreversible operations without confirmation. Installable per‑project or globally, it works entirely offline and is licensed under MIT.

A lightweight PyTorch library for differentiable 3D rotation manipulation, providing mappings between rotation representations and utilities for rotation regression.

A suite of rerankers featuring pointwise, pairwise, and listwise models to improve the accuracy of search and retrieval systems.

A Jax-based proof-of-concept implementation of Stable-Baselines3, providing high-performance reinforcement learning algorithms with a consistent API.

gmft is a lightweight, high-performance PDF table extraction tool that uses Microsoft's Table Transformer to convert PDF tables into structured formats like Pandas dataframes.

An open-source framework for high-performance medical image processing, neural network inference, and real-time visualization using CPUs and GPUs.

A multi-modal AI model that integrates Vision Transformer (ViT) and Mamba to efficiently process and interpret text, images, audio, and video data simultaneously.

SDAR (Synergy of Diffusion and AutoRegression) is an open‑source LLM family that blends autoregressive training with discrete diffusion decoding. The approach yields 2‑4× faster inference while matching AR model quality, especially on science‑reasoning benchmarks. The repository provides pretrained models (1.7 B‑30 B), a simple generate script, and two high‑performance inference engines (JetEngine and LMDeploy). Training/fine‑tuning scripts use LLaMA‑Factory, and the project is MIT‑licensed and actively maintained.

A Streamlit web app that uses LLMs (e.g., Qwen2.5) and BGE‑M3 embeddings to extract claims from news, retrieve web evidence via DuckDuckGo/SearXNG, and verify the claims through a three‑node pipeline. Supports Chinese, English, Japanese, Korean, multiple model providers, user accounts, history, and PDF reports.

An intent compiler for AI agent coding that transforms human requirements into verifiable Task Contracts to ensure AI-generated code is mechanically validated.

A local-first AI terminal operator that converts natural language intent into governed, reviewable, and reversible system actions with built-in safety guardrails.

A comprehensive iOS SDK for secure identification document scanning, offering automated data extraction from IDs and passports with an optional ready-to-use UI.

A multi-user agent platform that allows AI agents to execute tasks on private, non-publicly routable machines via an outbound connector or cloud sandboxes.

A deep-learning framework for automatic line art colorization that supports guidance via colored hints, text tags, or reference images.

A stand-alone implementation of NumPy dtype extensions for machine learning, providing support for bfloat16, 8-bit floats, and narrow integers.

A terminal session manager that allows AI agents to run, monitor, and interact with background dev servers and TUIs without blocking.

A private AI workspace for macOS that provides chat, dictation, and meeting notes while keeping data local and routing inference through a verifiable, privacy-preserving API.

A PyTorch implementation of the SLAYER framework for training multilayer spiking neural networks using backpropagation to learn synaptic weights and axonal delays.

A production-ready governance framework for enterprise AI agents that implements safety guardrails, compliance-as-code, IAM, and human-in-the-loop escalation.

An open-source machine learning framework for building contextual AI assistants and agents that automate text and voice-based conversations.

Ivy is a transpiler that allows developers to convert machine learning code, models, and libraries between frameworks such as PyTorch, TensorFlow, JAX, and NumPy.

An open-source face recognition library using deep neural networks to identify people from images and real-time video streams.

An autonomous ML agent that researches, writes, and ships machine learning code by integrating deeply with the Hugging Face ecosystem and cloud compute.

LMFlow is an extensible Python toolbox for fine‑tuning large language models. It supports full‑parameter training, LoRA, the memory‑efficient LISA algorithm, QLoRA quantization, and a suite of speed‑up tricks (FlashAttention‑2, DeepSpeed, vLLM, SGLang). After training, it provides simple scripts for chat‑style inference and optional Gradio/Flask deployment. The library is designed for Linux/Colab environments and offers optional extras for multimodal models, distributed reward‑model inference, and more.

A cross-platform open-source machine learning framework for .NET developers to build, train, and deploy custom models without needing Python or R.

torchtune is a PyTorch‑based library that provides ready‑to‑run recipes and configs for fine‑tuning, RL‑HF, distillation, quantization‑aware training, and inference of many modern LLMs (Llama, Gemma, Qwen, Phi, etc.). It abstracts distributed training, memory‑saving tricks, and device support behind a simple CLI (`tune`) and YAML files, integrating with Hugging Face, torch‑ao, FSDP2, and logging tools. Though development stopped in 2025, it remains a practical, open‑source toolkit for LLM experimentation.

A scikit-learn compatible neural network library that wraps PyTorch, enabling the use of scikit-learn's pipelines and grid search with PyTorch models.

pycorrector is a Python toolkit for Chinese text correction. It offers a fast KenLM‑based rule system and a suite of neural models (MacBERT, T5, ERNIE, GPT‑style LLMs, BART, etc.) that can be used out‑of‑the‑box or fine‑tuned. Install via pip, pick a model class, and call `correct()` or `correct_batch()` to get corrected sentences and error details. The repo includes evaluation scripts, demos, and command‑line batch processing, targeting use‑cases like input methods, OCR/ASR post‑processing, and search query cleaning.

A deep learning model that converts hand-drawn web mockups into HTML code using an image captioning architecture.