A Curated List of Vision-Language-Action (VLA) and World Action Models (WAM) Research and Beyond

PartFlow: two-stage image-conditioned 3D editing (inference code)

A framework and CLI toolkit for orchestrating teams of loosely-coupled AI agents.

[CVPR 2026 Highlight🔥] "NeAR: Coupled Neural Asset–Renderer Stack" Official Implementation

[CVPR' 26] Benchmarking PhD-Level Coding in 3D Geometric Computer Vision

Unified Map Prior Encoder for Mapping and Planning

[SIGGRAPH 2026] Official code of the paper "Relit-LiVE: Relight Video by Jointly Learning Environment Video".

[SIGGRAPH 2026 / TOG] Official code of the paper "UniVidX: A Unified Multimodal Framework for Versatile Video Generation via Diffusion Priors".

Implementation of Peter Shirley's Ray Tracing In One Weekend book using Vulkan and NVIDIA's RTX extension.

Conditional Memory via Scalable Lookup: A New Axis of Sparsity for Large Language Models

A real-time Gaussian Splatting implementation for SLAM systems, building upon the excellent work of MonoGS and Photo-SLAM.

[CORL 2025 Oral]One View, Many Worlds: Single-Image to 3D Object Meets Generative Domain Randomization for One-Shot 6D Pose Estimation.

First Latency-Aware Competitive LLM Agent Benchmark

QaLoF (Quantization-Aware LoRA Finetuning) with per-layer bit-width control, cyclic precision training, and adversarial robustness for GPT-2 models

Main Web Site (Online Books)

Cosmos-Reason1 models understand the physical common sense and generate appropriate embodied decisions in natural language through long chain-of-thought reasoning processes.

[CVPR 2025 Highlight] GEN3C: 3D-Informed World-Consistent Video Generation with Precise Camera Control

Original implementation of "Radiant Foam: Real-Time Differentiable Ray Tracing"

[CVPR 2025] Sparse Voxels Rasterization: Real-time High-fidelity Radiance Field Rendering

Fast Multimodal LLM on Mobile Devices

Codes for Switch-NeRF (ICLR 2023)

The best OSS video generation models, created by Genmo

Burn is a next generation tensor library and Deep Learning Framework that doesn't compromise on flexibility, efficiency and portability.

3D Reconstruction for all

Next-Token Prediction is All You Need