Zyan

PixWorld. PixWorld: Unifying 3D Scene Generation and Reconstruction in Pixel Space

Syn4D. [ECCV 2026] Syn4D: A Multiview Synthetic 4D Dataset

GenRecon. Python

Lift4D. Lift4D: Harmonizing Single-View 3D Estimation for 4D Reconstruction In-the-Wild

world-tracing. Office inference code for World Tracing (object/scene/dynamic). Live demos: https://haoz19.github.io/world-tracing-page/

ChordEdit. [CVPR 2026 Oral] Official implementation for ChordEdit: One-Step Low-Energy Transport for Image Editing

PhotoFlow. PhotoFlow: Agentic 3D Virtual Photography Missions

learning-convex-decomposition. Python

GalaxeaVLA. Galaxea's open-source VLA repository

TripoSplat. TripoSplat converts a single 2D image into high-quality and variable number of 3D Gaussians, developed by TripoAI.

vgg-ttt. [CVPR'26] Official code for the paper "VGG-T³: Offline Feed-Forward 3D Reconstruction at Scale"

symphony. Symphony turns project work into isolated, autonomous implementation runs, allowing teams to manage work instead of supervising coding agents.

streaming. A Data Streaming Library for Efficient Neural Network Training

RAEv2. Official Implemenation for RAEv2: Improved Baselines with Representation Autoencoders

NoPo4D. Official implementation of No Pose, No Problem in 4D: Feed-Forward Dynamic Gaussians from Unposed Multi-View Videos

vggt-omega. [CVPR 2026 Oral] VGGT Omega

ml-lito. [ICLR 2026] LiTo: Surface Light Field Tokenization

karing. Simple & Powerful proxy utility, Support routing rules for clash/sing-box

taste-skill. Taste-Skill - gives your AI good taste. stops the AI from generating boring, generic slop

OmniFit. [ECCV 2026] OmniFit: Multi-modal 3D Body Fitting via Scale-agnostic Dense Landmark Prediction

FD-Loss. Python

STream3R. Dynamic 3D Foundation Model using Causal Transformer. [ICLR 2026]

powerfoam. Python

MV-SAM3D. SAM 3D Objects with Multi-view Images

Scal3R. [CVPR 2026 (Highlight)] Scal3R: Scalable Test-Time Training for Large-Scale 3D Reconstruction

Hy3-preview. Hy3 preview (295B A21B), a leading reasoning and agent model in its size, with great cost efficiency

4C4D. [CVPR 2026] 4C4D: 4 Camera 4D Gaussian Splatting

WorldStereo. [CVPR 2026] WorldStereo: Bridging Camera-Guided Video Generation and Scene Reconstruction via 3D Geometric Memories (WorldExpand of HY-World 2.0)

WeClone. 🚀 One-stop solution for creating your AI twin from chat history 💡 Fine-tune LLMs with your chat logs to capture your unique style, then bind to a chatbot to bring your digital self to life.

HY-World-2.0. HY-World 2.0: A Multi-Modal World Model for Reconstructing, Generating, and Simulating 3D Worlds

graphify. AI coding assistant skill (Claude Code, Codex, OpenCode, Cursor, Gemini CLI, and more). Turn any folder of code, SQL schemas, R scripts, shell scripts, docs, papers, images, or videos into a queryable knowledge graph. App code + database schema + infrastructure in one graph.

cuVSLAM. cuVSLAM: CUDA-Accelerated Visual Odometry and Mapping

LoMa. [ECCV 2026] LoMa: Local Feature Matching Revisited

neural-harmonic-textures. Neural Harmonic Textures for High-Quality Primitive Based Neural Reconstruction

boxer. Code for the Boxer research paper

WildDet3D. Allen Institute for AI: WildDet3D: Scaling Promptable 3D Detection in the Wild

OmniRoam. [SIGGRAPH 2026] OmniRoam: World Wandering via Long-Horizon Panoramic Video Generation

AlayaRenderer. Generative World Renderer: an AI-native Renderer for Games and Virtual Worlds.

fullcircle. Implementation of "FullCircle: Effortless 3D Reconstruction from Casual 360° Captures"

build_gpu_colmap. PowerShell

claw-code. An agent-managed museum exhibit, built in Rust with Gajae-Code / LazyCodex — developed and maintained with no human intervention.

resplat. ReSplat: Learning Recurrent Gaussian Splatting

AutoGaze. AutoGaze automatically removes redundant patches in a video, reducing #tokens in ViT/MLLM by 4x-100x.

paperclip. The open-source app everyone uses to manage agents at work

gram-newton-schulz. Fast Polar Decomposition for Muon

pretext. Fast, accurate & comprehensive text measurement & layout

le-wm. Official code base for LeWorldModel: Stable End-to-End Joint-Embedding Predictive Architecture from Pixels

modular. The Modular Platform (includes MAX & Mojo)

inspatio-world. Python

clihub. The App Store for AI Agents — discover, install, and manage 100+ CLI tools

fvdb-reality-capture. Jupyter Notebook

video_to_world. Our method reconstructs 3D worlds from video diffusion models using non-rigid alignment to resolve inherent 3D inconsistencies in the generated sequences.

fvdb-core. Python

parameter-golf. Train the smallest LM you can that fits in 16MB. Best model wins!

CLI-Anything. "CLI-Anything: Making ALL Software Agent-Native" -- CLI-Hub: https://clianything.cc/

clihub. The App Store for AI Agents — discover, install, and manage 100+ CLI tools via structured JSON.

awesome-cli-apps. 🖥 📊 🕹 🛠 A curated list of command line apps

CubeComposer. [CVPR 2026] Spatio-Temporal Autoregressive 4K 360° Video Generation from Perspective Video

argus-code. [ICCV 2025] Official repository of the paper "Beyond the Frame: Generating 360° Panoramic Videos from Perspective Videos"

AutoResearchClaw. Fully autonomous & self-evolving research from idea to paper. Chat an Idea. Get a Paper. 🦞

Geometry-Grounded-Gaussian-Splatting. Geometry-Grounded Gaussian Splatting

Attention-Residuals.

blunt. BLUNT — Basic Lifting and UNprojection Tool. Turn any photo into a 3D Gaussian Splat in seconds.

gstack. Use Garry Tan's exact Claude Code setup: 23 opinionated tools that serve as CEO, Designer, Eng Manager, Release Manager, Doc Engineer, and QA

Fast-FoundationStereoPhysics. A standlone demo from ManiDreams: Fast-FoundationStereo + SAM2 + Newton, zero-shot real-time simulation-based world model

Helios. Helios: Real Real-Time Long Video Generation Model

packuv. PackUV: Packed Gaussian UV Maps for 4D Volumetric Video

3dgeer. [ICLR 2026] The official implementation associated with the paper "3DGEER: 3D Gaussian Rendering Made Exact and Efficient for Generic Cameras"

robot_descriptions.py. Access 185+ robot descriptions from the main Python robotics frameworks

LoGeR. Reimplementation of LoGeR: Long-Context Geometric Reconstruction with Hybrid Memory

EgoX. Code for "EgoX: Egocentric Video Generation from a Single Exocentric Video"

SpatialVID. [CVPR 2026] SpatialVID: A Large-Scale Video Dataset with Spatial Annotations

Self-Flow. [ICML'26] Code and website for Self-Flow: Self-Supervised Flow Matching for Scalable Multi-Modal Synthesis

worldfm. Python

nanotron. Minimalistic large language model 3D-parallelism training

autoresearch. AI agents running research on single-GPU nanochat training automatically

ZipMap. [CVPR 2026] ZipMap: Linear-Time Stateful 3D Reconstruction via Test-Time Training

interactive_world_sim. [RSS 2026] Interactive World Simulator for Robot Policy Training and Evaluation

claude-code-best-practice. from vibe coding to agentic engineering - practice makes claude perfect

Track4World. [ECCV 2026] Track4World: Feedforward World-centric Dense 3D Tracking of All Pixels

Utonia. [ICML'26] Official repository of Utonia: Toward One Encoder for All Point Clouds

LichtFeld-Studio. Train, inspect, edit, automate, and export 3D Gaussian Splatting scenes from a single native application.

3dgrut. Ray tracing and hybrid rasterization of Gaussian particles

rapier. 2D and 3D physics engines focused on performance.

spark-physics. JavaScript

character-controller-sample-project. A 3D character controller built with React Three Fiber, Rapier physics.

context7. Context7 Platform -- Up-to-date code documentation for LLMs and AI code editors

skills. Public repository for Agent Skills

Scrapling. 🕷️ An adaptive Web Scraping framework that handles everything from a single request to a full-scale crawl!

PointWorld. PointWorld: Scaling 3D World Models for In-The-Wild Robotic Manipulation

dino_wm. Python

ratatui. A Rust crate for cooking up terminal user interfaces (TUIs) 👨‍🍳🐀 https://ratatui.rs

tortuise. Gaussian splats in your terminal.

openclaw-wechat. 让 OpenClaw稳定的连上你的个人微信

Reward-Forcing. [CVPR 2026 Highlight] Reward Forcing: Efficient Streaming Video Generation with Rewarded Distribution Matching Distillation

Magic-World. official code for "magicworld: towards long-horizon stability for interactive video world exploration"

Uni3C. Uni3C: Unifying Precisely 3D-Enhanced Camera and Human Motion Controls for Video Generation [Siggraph Asian 2025]

tttLRM. [CVPR 2026 Highlight] tttLRM: Test-Time Training for Long Context and Autoregressive 3D Reconstruction

GR00T-WholeBodyControl. Welcome to GR00T Whole-Body Control (WBC)! This is a unified platform for developing and deploying advanced humanoid controllers. This includes: Decoupled WBC models used in NVIDIA Isaac-Gr00t, Gr00t N1.5 and N1.6 and GEAR-SONIC

YoNoSplat. [ICLR'26] YoNoSplat: You Only Need One Model for Feedforward 3D Gaussian Splatting

Any4D. Any4D: Unified Feed-Forward Metric 4D Reconstruction

prompts.chat. f.k.a. Awesome ChatGPT Prompts. Share, discover, and collect prompts from the community. Free and open source — self-host for your organization with complete privacy.

DreamDojo. Official Codebase for "DreamDojo: A Generalist Robot World Model from Large-Scale Human Videos" (ICML 2026)

meta-wearables-dat-ios. Meta Wearables Device Access Toolkit for iOS

SR-3D. [ICLR'26] This repository is the implementation of "3D Aware Region Prompted Vision Language Model"

fantasy-world. [ICLR 2026] FantasyWorld: Geometry-Consistent World Modeling via Unified Video and 3D Prediction

claude-skills. 66 Specialized Skills for Full-Stack Developers. Transform Claude Code into your expert pair programmer.

zeroclaw. Fast, small, and fully autonomous AI personal assistant infrastructure, any OS, any platform — deploy anywhere, swap anything 🦀

picoclaw. Tiny, Fast, and Deployable anywhere — automate the mundane, unleash your creativity

VLA-JEPA. [ECCV 2026] VLA-JEPA: Enhancing Vision-Language-Action Model with Latent World Model

MMaDA. MMaDA - Open-Sourced Multimodal Large Diffusion Language Models (dLLMs with block diffusion, mixed-CoT, unified RL)

drifting-model. Personal PyTorch implementation of "Generative Modeling via Drifting" with Claude

spz. File format for 3D Gaussian splats. About 10x smaller than the PLY equivalent with virtually no perceptible loss in visual quality. Offered as open source by Niantic Labs. More details at https://scaniverse.com/spz #3dgaussiansplats #gaussiansplatting

Rolling-Sink. Official implementation of "Rolling Sink: Bridging Limited-Horizon Training and Open-Ended Testing in Autoregressive Video Diffusion"

MotionCrafter. [CVPR 2026 Highlight🔥] MotionCrafter: Dense Geometry and Motion Reconstruction with a 4D VAE

PLAS. Parallel Linear Assignment Sorting - Sorting multi-dimensional data into locally smooth 2D grids.

Self-Organizing-Gaussians. [ECCV '24] Compressing 3D Gaussian Splats by placing their parameters into a 2D grid with local smoothness

3DGen-Playground. 3D generation made easy!

DiffusionNFT. [ICLR 2026 Oral] DiffusionNFT: Online Diffusion Reinforcement with Forward Process

DeFi. [ICLR 2026] Disentangled Robot Learning via Separate Forward and Inverse Dynamics Pretraining

VisionClaw. Real-time AI assistant for Meta Ray-Ban smart glasses -- voice + vision + agentic actions via Gemini Live and OpenClaw

Awesome-3D-Scene-Generation. A curated list of awesome 3D scene generation papers. (arXiv 2505.05474)

dreamzero. Code to pretrain, fine-tune, and evaluate DreamZero and run sim & real-world evals

pMF. Official Implementation of pMF https://arxiv.org/abs/2601.22158

eb_jepa. An open source library designed to provide community examples of Joint Embedding Predictive Architectures (JEPAs). It contains code and examples for learning representations from images, video, and action-conditioned video, as well as planning using JEPA-based models.

Causal-Forcing. [ICML 2026] Official codebase for "Causal Forcing: Autoregressive Diffusion Distillation Done Right for High-Quality Real-Time Interactive Video Generation" & Causal Forcing++

FastGen. NVIDIA FastGen: Fast Generation from Diffusion Models

SceneMaker. [CVPR 2026] Implementation of paper "SceneMaker: Open-set 3D Scene Generation with Decoupled De-occlusion and Pose Estimation Model"

kaolin. A PyTorch Library for Accelerating 3D Deep Learning Research

TRELLIS.2. Native and Compact Structured Latents for 3D Generation

lingbot-world. Advancing Open-source World Models

actionmesh. 🎬ActionMesh: A fast video to animated mesh model with unprecedented quality. Generate animated mesh seamlessly importable into any 3D software in less than a minute.

Qwen-Image-Layered. Qwen-Image-Layered: Layered Decomposition for Inherent Editablity

MetalSplatter. Render Gaussian Splats using Metal on Apple platforms (iOS/iPhone/iPad, macOS, and visionOS)

ppisp. Physically-Plausible Image Signal Processing (PPISP) for Radiance Field Reconstruction

openclaw. Your own personal AI assistant. Any OS. Any Platform. The lobster way. 🦞

VIGA. VIGA: Vision-as-Inverse-Graphics Agent

splatkit. Splatkit enables you to create and share high quality dynamic gaussian splats that build on the latest research.

representations4d. Jupyter Notebook

Diffusion-Explorer. Interactive visualizations of the geometric intuition behind diffusion models.

dreamer4. Unofficial implementation of the Dreamer 4 world model in PyTorch.

Fixer. Official Repository for the NVIDIA Fixer model

VerseCrafter. VerseCrafter: Dynamic Realistic Video World Model with 4D Geometric Control

ralph-wiggum-cursor. Cursor CLI implementation of Geoffrey Huntley's Ralph Wiggum autonomous iteration technique with deliberate context management

Perspective-Effect-with-Head-Tracking-of-MediaPipe. HTML

jepa-wms. Code, data and weights for the paper **What drives success in physical planning with Joint-Embedding Predictive World Models?**

DiT360. [CVPR 2026] Official implementation of "DiT360: High-Fidelity Panoramic Image Generation via Hybrid Training".

torchcodec. PyTorch media decoding and encoding

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

awesome-spatial-intelligence. 🌐 Forging Spatial Intelligence: A Roadmap of Multi-Modal Data Pre-Training for Autonomous Systems

TileGym. Helpful kernel tutorials, examples and SKILLs for tile-based GPU programming

OMG. Optimized Minimal 3D Gaussian Splatting

gsplat.js. JavaScript Gaussian Splatting library.

Fast-FoundationStereo. [CVPR 2026] Fast-FoundationStereo: Real-Time Zero-Shot Stereo Matching

TurboDiffusion. TurboDiffusion: 100–200× Acceleration for Video Diffusion Models

UCPE. 📷 [CVPR'26] Camera-controlled text-to-video generation, now with intrinsics, distortion and orientation control!

HY-WorldPlay. HY-World 1.5: A Systematic Framework for Interactive World Modeling with Real-Time Latency and Geometric Consistency

ml-sharp. Sharp Monocular View Synthesis in Less Than a Second

metamorph. Code for MetaMorph Multimodal Understanding and Generation via Instruction Tuning

VGGT-X. VGGT-X: When VGGT Meets Dense Novel View Synthesis

OMG4. Python

E-RayZer. [CVPR 2026] "E-RayZer: Self-supervised 3D Reconstruction as Spatial Visual Pre-training" official implementation.

TensorRT. NVIDIA® TensorRT™ is an SDK for high-performance deep learning inference on NVIDIA GPUs. This repository contains the open source components of TensorRT.

cupy. NumPy & SciPy for GPU

cutile-python. cuTile is a programming model for writing parallel kernels for NVIDIA GPUs

cutlass. CUDA Templates and Python DSLs for High-Performance Linear Algebra

visionary. Visionary: The World Model Carrier Built on WebGPU-Powered Gaussian Splatting Platform

sam3. The repository provides code for running inference and finetuning with the Meta Segment Anything Model 3 (SAM 3), links for downloading the trained model checkpoints, and example notebooks that show how to use the model.

VGGT4D. The official implementation of the paper “VGGT4D: Mining Motion Cues in Visual Geometry Transformers for 4D Scene Reconstruction.”

Z-Image. Python

world-in-world. "World Models in a Closed-Loop World" (ICLR'26 Oral)

scope. A tool for running and customizing real-time, interactive generative AI pipelines and models

physctrl. [NeurIPS 2025] PhysCtrl: Generative Physics for Controllable and Physics-Grounded Video Generation

RayZer. Code for ICCV'2025 (Best student paper honorable mention) "RayZer: A Self-supervised Large View Synthesis Model"

Awesome-Controllable-Video-Generation. [ArXiv 2025] A survey about controllable video generation: This repo is the official awesome of "Controllable video generation: A survey"

xfactor-nvs. Public code for XFactor: Introduces the first geometry-free model to achieve true self-supervised / pose-free Novel View Synthesis (NVS) by learning transferable latent camera pose representations.

sim-evals. A simulation evaluation platform for DROID

Online-3DGS-Monocular. Code repo for the SIGGRAPH paper "Monocular Online Reconstruction with Enhanced Detail Preservation". Project page https//poiw.github.io/MODP/index.html

LocateAnything3D.

GNS. Official code implementation of paper "Gradient-Driven Natural Selection for Compact 3D Gaussian Splatting".

Muon. Muon is an optimizer for hidden layers in neural networks

FastGS. [CVPR 2026 Highlight, the CVPR Compute Gold Star] Offical code for "FastGS: Training 3D Gaussian Splatting in 100 Seconds"

splat-transform. CLI tool and library for 3D Gaussian splat processing and conversion

CoMe. [CVPR 26] Release repo of our work "Co-Me: Confidence-Guided Token Merging for Visual Geometric Transformers"

IGGT_official. [ICLR'26] IGGT: Instance-Grounded Geometry Transformer for Semantic 3D Reconstruction

cambrian-s. Cambrian-S: Towards Spatial Supersensing in Video

JiT. PyTorch implementation of JiT https://arxiv.org/abs/2511.13720

Pi-Long. Code implementation of Pi-Long

VGGT-Long. Official implement of VGGT-Long

sam-3d-objects. SAM 3D Objects

GeoVista. Official repo for "GeoVista: Web-Augmented Agentic Visual Reasoning for Geolocalization"

MeshCoder. Jupyter Notebook

DreamCube. [ICCV 2025] Official implementation of the paper "DreamCube: 3D Panorama Generation via Multi-plane Synchronization".

Depth-Anything-3. Depth Anything 3

projectaria_gen2_pilot_dataset. Data loader and visualization tool for Project Aria Gen2 pilot dataset.

DA-2. Official implementation of "DA²: Depth Anything in Any Direction"

sglang. SGLang is a high-performance serving framework for large language models and multimodal models.

rcm. rCM & Causal-rCM: Leading and Unified Algorithms/Infrastructures for Bidirectional/Autoregressive Video Diffusion Distillation at Scale

lejepa. Python

UniPixel. 🔮 UniPixel: Unified Object Referring and Segmentation for Pixel-Level Visual Reasoning (NeurIPS 2025)