A framework for high-quality real-time interactive video generation using Causal ODE or Causal Consistency Distillation to enable efficient few-step autoregressive diffusion.

An agentic skill that gives LLMs access to Ghidra via a JSON-based CLI and background daemon, allowing AI to autonomously perform reverse engineering tasks.

VBench is an open‑source benchmark suite that automatically evaluates video‑generation models across dozens of quality dimensions (technical, aesthetic, and trustworthiness). It includes prompt collections, metric pipelines, human‑aligned scores, and public leaderboards for T2V, I2V, long‑video, and intrinsic‑faithfulness evaluation.

GigaBrain‑0.7 is an open‑source, large‑scale vision‑language‑action model for robots. It is pretrained on 37 k h of heterogeneous embodied data, supports eight robot embodiments, and ships with training scripts, benchmark suites (VLM‑Benchmark, RoboColiseum, RoboTwin 2.0, EBench), and ready‑to‑run inference servers that integrate with ROS.

A control-room framework for language-agent societies that allows users to observe, pause, and intervene in AI agent simulations in real-time.

LeKiwi is a low-cost mobile manipulator featuring a holonomic drive and a robotic arm, designed for accessible robotics research and data collection.

A plugin marketplace and MCP server collection that connects AI assistants like GitHub Copilot to Microsoft 365 data for natural language querying of emails, meetings, and documents.

A collection of low-level agentic design patterns and skills for AI agents that prioritize subtraction and refinement to prevent AI-generated bloat and repository rot.

A collection of database-focused skills for AI coding assistants to provide expert guidance on MySQL, Postgres, Neki, and Vitess.

A tool that generates custom Gridfinity 3D-printable bins from photos of tools using AI-powered silhouette tracing and scale calibration.

A native desktop AI assistant that provides a unified interface for managing multiple AI CLI tools, git worktrees, and project sessions without vendor lock-in.

A toolkit for training neural networks to model the behavior of guitar amplifiers and audio equipment, exporting them for real-time use.

A toolkit for giving AI agents crypto wallets and onchain interaction capabilities, enabling them to to handle payments and monetize their services.

A Go SDK that provides a thin, idiomatic wrapper around Anthropic’s Claude API, enabling Go developers to call Claude models (e.g., Claude Opus 4.6) directly from their applications. Install with `go get`, create a client with an API key, and use the `Messages.New` method to send prompts and receive responses.

A local AI agent framework for iOS that runs LLMs on-device to privately manage calendar, health data, and device tasks via natural language.

ComfyUI‑Trellis2 is a custom‑node pack that integrates Microsoft’s TRELLIS.2 3‑D diffusion model into the visual programming UI ComfyUI. It provides dozens of nodes for generating, refining, texturing, and rendering meshes from single or multi‑view images (or video). Installation requires Windows, Python 3.11, PyTorch 2.7/2.8 with CUDA, and the Facebook DINOv3 checkpoint. The repo ships pre‑built wheels for native extensions and includes example workflows. It’s an active, genuine AI/ML project aimed at 3‑D generation.

An AI-powered mistake management system that helps students organize, analyze, and review wrong answers through automated tagging and similar-problem generation.

LatticeDB is an embedded, single‑file property‑graph database that also supports HNSW vector similarity and BM25 full‑text search. It offers a unified Cypher query language, ACID transactions, durable change‑feeds, and bindings for Python, TypeScript, Go, and Java. Benchmarks show sub‑microsecond node lookups, sub‑millisecond 10‑NN searches on 1 M vectors, and microsecond full‑text queries, all without a server. Ideal for local knowledge graphs, agent memory, and RAG pipelines, but not suited for multi‑writer or distributed workloads.

An MCP server providing AI agents with real-time global intelligence across 30+ domains, featuring 132 tools and a Qdrant vector store for semantic search.

An automated multi-agent system that generates institutional-grade, citable research reports in 19 languages using a four-stage search and verification pipeline.

A desktop runtime middleware that allows users to access and operate Codex AI coding tools remotely via a web browser from any device.

An offline, privacy-focused Android document scanner that uses on-device ML and OCR to create searchable PDFs for self-hosted workflows.

A set of Codex skills that transforms raw business materials into professional consulting slides and converts image-based presentations into editable PowerPoint files.

A local-first audio and video transcription tool that converts recordings into structured Markdown notes and enables AI-powered Q&A over the content.

VeOmni is a versatile framework for scaling the pre-training and post-training of single- and multi-modal models across various hardware accelerators using a modular, trainer-free design.

A tiny, self-hosted notification inbox for developers that pushes app events and errors directly to an iOS app via APNs without third-party relays.

A native macOS application that manages a local proxy server to centralize AI coding assistant accounts, track quotas, and automate agent configuration.

A UI framework providing enterprise-level components and SDKs for building AI-driven interfaces, specifically optimized for LLM streaming and dynamic interaction.

An open-source, AI-powered CRM system that provides self-hosted customer relationship management with integrated AI assistants and BI data visualization.

A framework for building stateful, long-running AI agents and collaborative apps using lightweight actors with co-located in-memory state and durable execution.

A Rust-native, drop-in replacement for Apache Spark that provides significantly faster execution and lower infrastructure costs while maintaining full compatibility with Spark SQL and the DataFrame API.

A Claude Code plugin that automates the full album production pipeline, from concept and research to lyrics, AI music prompts, and audio mastering.

An AI video analysis tool that reverse-engineers existing videos to extract prompts and scripts, helping creators replicate and iterate on successful AI video styles.

ReLE is a comprehensive benchmark for Chinese LLMs, providing detailed leaderboards across 300+ dimensions and a 2-million-entry defect library to help users select and improve models.

A local AI SDK and model provider that enables running LLMs, speech, vision, and generative AI workloads on any device without cloud dependencies.

An open source full-stack framework for building and deployable chatbots and conversational apps using React, Serverless, and Tensorflow.js.

A multi-agent workspace that transforms existing AI CLI tools into a unified orchestration hub for parallel execution and automated workflows.

A memory-centric AI research agent that automates the full scientific research lifecycle, from literature ingestion and idea generation to experimental execution and manuscript writing.

An RLM-based agent optimizer that analyzes production execution traces to identify systemic failure modes and provide recommendations for improving agent harnesses.

A CLI for spec-driven development that turns product intent into a governed AI coding workflow with repo-native tracking and git worktree isolation.

A pure-Python reference implementation of the CuTe layout-and-tensor algebra used in CUTLASS 3.x, enabling the prototyping and analysis of GPU memory layouts without a GPU.

A collection of 11 autonomous SEO skills for Claude and other AI agents that perform live research, competitor analysis, and content auditing without requiring manual data exports.

A distributed GPU-accelerated framework for scalable evolutionary computation that provides over 50 algorithms and 100 benchmark problems compatible with PyTorch.

A comprehensive suite of 457 custom nodes for ComfyUI that adds advanced image processing, 3D scene rendering, and complex logic flow to generative AI workflows.

An autonomous multi-model AI agent swarm for offensive and defensive security automation, designed to solve CTFs and perform penetration testing by coordinating diverse LLMs via a shared blackboard.

Codex‑Manager is a Rust/Tauri desktop and service app that lets you pool, manage, and route OpenAI‑style API keys (Codex, Gemini, Claude, etc.). It provides UI‑driven account import/export, quota tracking, automated reset reminders, a local OpenAI‑compatible gateway, model routing, and a plugin/skill ecosystem.

A visual fiducial system for robotics that detects unique markers in images to provide precise identification and 3D pose estimation.

A platform for reproducible world model research and evaluation that provides a unified interface for data collection, training, and model-predictive control evaluation.