A self-hosted AI agent that runs locally on your machine, providing a native desktop app and integration with multiple messaging platforms like Discord and Slack.

A LangGraph template for implementing a ReAct agent that iteratively reasons and executes tool-based actions to solve complex user queries.

A prompt-native programming system that treats .prompt files as the source of truth and generates implementation code as a reviewable artifact.

A neural texture compression SDK that compresses multiple PBR texture channels into a single neural set to reduce VRAM and bandwidth usage in 3D rendering.

A collection of advanced scenario-driven demos showing how to integrate OpenAI ChatKit with FastAPI and React to create AI agents that synchronize with real-time UI state and interactive widgets.

A private, low-latency voice assistant that runs locally on your hardware using a stack of speech recognition, LLM, and speech generation models.

A deterministic MoE training megakernel for NVIDIA Blackwell NVL72s that fuses computation and communication to eliminate CPU-GPU synchronization and accelerate training.

CC‑Panes is an open‑source desktop app (Windows/macOS/Linux) that lets you run, organise and coordinate multiple AI‑coding CLI agents in parallel. It provides workspace‑level dashboards, reusable launch profiles, an MCP‑based orchestration layer, integrated terminal splits, Git/history tools, planning surfaces, and web/Android access—all built with Tauri, xterm.js and Rust.

EgoVerse is a framework and dataset for robot learning from egocentric human demonstrations, providing tools for data processing, training, and evaluation.

An autonomous, AI-driven penetration testing framework that orchestrates over 72 security tools to automate reconnaissance, exploitation, and reporting.

VideoChat3 is a 4B generalist video MLLM designed for efficient video understanding, covering everything from subtle motion and long-form reasoning to live stream proactive responses.

Custom ComfyUI nodes that integrate Depth Anything V3 for high-quality depth estimation, 3D point cloud reconstruction, and consistent video depth mapping.

An open-source parallel coding agent harness that provides a single web portal to run and control multiple AI coding agents simultaneously.

A cross-platform note-taking app with support for Markdown, LaTeX, and Mermaid diagrams, featuring an optional AI co-writer that connects to various LLM providers.

An MCP server that connects AI assistants to South Korea's MOLIT real estate API to provide live transaction data and financial investment simulations.

A modular suite of AI agent skills for self-media content production, covering everything from trend research and strategy to platform-native copywriting and publishing.

An MCP server that provides AI coding agents with structured, full-content web search results from technical sites like GitHub and StackOverflow to improve code generation quality.

An AI-powered smartphone automation assistant that uses vision language models to execute natural language commands on Android via accessibility services.

A Python CLI framework that stores threat‑hunting investigations as markdown using the LOCK pattern, makes them searchable, and optionally lets LLM agents generate research, run queries, and automate parts of the hunt.

An AI coding assistant that provides persistent workspaces with connected data and the ability to build and deploy applications directly from the chat environment.

FlagTree is a unified open-source compiler for multiple AI chips that enables a "develop once, run anywhere" workflow by unifying various Triton-based backends.

A local-first AI filmmaking workspace that integrates coding agents with a visual canvas and a unified API for generating images, video, and voice.

A security skills framework and ruleset that embeds secure-by-default practices into AI coding agents to prevent vulnerabilities during code generation and review.

An agentic AI framework for designing and orchestrating multi-agent systems with distributed execution environments, hybrid memory, and built-in safety guards.

A developer framework for integrating machine learning into graphics applications, enabling the use of neural networks to represent shaders and textures.

A blazing-fast LLM inference engine written in pure Rust that eliminates Python/PyTorch dependencies to enable high-throughput, production-ready model serving on consumer and professional GPUs.

Lance is a 3B parameter unified multimodal model that integrates image and video understanding, generation, and editing into a single framework.

NVIDIA Cosmos‑Predict 2.5 is an open‑source diffusion model that turns text, images or video into future‑state video predictions. It supports 2 B/14 B checkpoints, robot‑ and autonomous‑vehicle‑specific variants, LoRA fine‑tuning, distillation, and integrates with Hugging Face Diffusers. Designed for physical AI (robotics, AV simulation, video analytics), it offers detailed docs, a cookbook of post‑training recipes, and is licensed under Apache 2.0 (code) and NVIDIA Open Model License (weights).

An AI-native library and MCP for end-to-end testing that allows users to perform actions and verify assertions in applications using natural language via Appium, Playwright, or Selenium.

A modular C++ and ROS 2 framework for robotics localization and SLAM, enabling configurable pipelines for LiDAR odometry and high-precision mapping.

A high-performance physics-enabled 3D simulator for embodied AI research, providing fast rendering and rigid-body dynamics for training agents in realistic 3D environments.

Rust bindings for the PyTorch C++ API (libtorch), enabling the use of PyTorch's tensor operations and deep learning tools in Rust.

A pure Rust implementation of TFHE that enables boolean and integer arithmetics to be performed directly on encrypted data without decryption.

A universal Japanese OCR popup dictionary that enables instant word lookups from any on-screen content, including games, manga, and videos.

An open-source Python framework for real-time data engineering and machine learning on Apache Kafka streams, providing a tabular DataFrame API for reliable ETL pipelines.

A high-fidelity navigation simulator that extends Habitat-Sim with 3D Gaussian Splatting to provide photo-realistic scenes and dynamic humanoid avatars for embodied AI research.

StreetCrafter is a framework for street view synthesis that uses controllable video diffusion models and 3D Gaussian Splatting to generate realistic videos from novel trajectories.

Manhattan‑SDF is a CVPR‑2022 research code that learns a neural signed‑distance‑function for indoor 3‑D reconstruction, adding a Manhattan‑world (axis‑aligned) regulariser to improve geometry quality. It provides conda‑based installation, scripts for training on ScanNet (or custom data), mesh extraction, and evaluation against many baselines.

A Blender add-on that integrates Stable Diffusion to generate seamless textures, concept art, and image assets directly within the 3D workspace.

A LiDAR filtering method based on cloth simulation that separates ground points from non-ground points in airborne point clouds.

A Python library for alpha matting that estimates the transparency of foreground objects using trimaps to enable precise background removal and composition.

A high-performance Python library for graph and hypergraph analysis that provides GPU acceleration and tools for higher-order network learning.

A serverless framework to deploy a customizable ChatGPT-style Telegram bot using Cloudflare Workers, supporting multiple AI providers and models.

A queryable knowledge base for video and photo archives that generates plain-text AI descriptions, transcripts, and metadata sidecars for every file.

A reliable benchmarking suite for time-series anomaly detection that provides curated datasets, a reliable evaluation metric (VUS-PR), and a unified framework to compare 40 different detection algorithms.

A Python-based multirotor simulation environment that models aerodynamic wrenches and wind to support research in UAV control, estimation, and reinforcement learning.

An Emacs plugin for GitHub Copilot that provides inline completions, an interactive chat interface, and agentic capabilities for code editing and review.

A lightweight, open-source desktop AI agent built in Rust that allows users to bring their own LLM keys to automate file and document management tasks locally and securely.