📹 A more flexible framework that can generate videos at any resolution and creates videos from images.

Wan: Open and Advanced Large-Scale Video Generative Models

High quality training free inpaint for every stable diffusion model. Supports ComfyUI

[NeurIPS 2025] Wan-Move: Motion-controllable Video Generation via Latent Trajectory Guidance

Local persistent memory store for LLM applications including claude desktop, github copilot, codex, antigravity, etc.

Depth Anything 3

SQL Native Memory Layer for LLMs, AI Agents & Multi-Agent Systems

An open-source, GPU-accelerated physics simulation engine built upon NVIDIA Warp, specifically targeting roboticists and simulation researchers.

A Python framework for accelerated simulation, data generation and spatial computing.

A multi-platform proxy client based on ClashMeta,simple and easy to use, open-source and ad-free.

A fully open source biomolecular structure prediction model based on AlphaFold3

Fast and Universal 3D reconstruction model for versatile tasks

High-throughput tensor loading for PyTorch

A lightweight suite of motion imitation methods for training controllers.

On-device TTS model by Neuphonic

[ ICCV 2025 ] FaceXFormer: A Unified Transformer for Facial Analysis

[AAAI 25] SegFace: Face Segmentation of Long-tail classes

BiSeNet: Bilateral Segmentation Network for Real-time Semantic Segmentation | Face Parsing | In PyTorch >> ONNX Runtime Inference

(CVPR 2023) CelebV-Text: A Large-Scale Facial Text-Video Dataset

Audio playback and capture library written in C, in a single source file.

Large multi-modal models (L3M) pre-training.

The open-source CapCut alternative

The Robot Operating System, is a meta operating system for robots.

Unified framework for robot learning built on NVIDIA Isaac Sim

Official implementation of SVFR.

SpatialVID: A Large-Scale Video Dataset with Spatial Annotations