Stars
[Tech Report] Context Scaling: Scaling Properties of Text Conditioning in Visual Generation
GPT-6 Astra for embodied AI and robotics.
RPent: Agentic Infrastructure for the Physical World
ViPE: Video Pose Engine for Geometric 3D Perception
[ICML'26] Code and website for Self-Flow: Self-Supervised Flow Matching for Scalable Multi-Modal Synthesis
Scaling Mixture-of-Experts Video Pretraining for Embodied Intelligence
[NeurIPS 2026] Next Forcing: Causal World Modeling with Multi-Chunk Prediction (MCP)
Infinite Worlds with Versatile Interactions
[CoRL 2026] ReactiveBFM: Reactive Closed-Loop Motion Planning Towards Universal Humanoid Whole-Body Control
Official Python inference and LoRA trainer package for the LTX-2 audio–video generative model.
Code for paper "ClothTransformer: Unified Latent-Space Transformers for Scalable Cloth Simulation (NeurIPS 2026)"
A 3B-active-parameter native unified multimodal model for image and video understanding, generation, and editing.
Implementation for paper "Forcing-KV: Hybrid KV Cache Compression for Efficient Autoregressive Video Diffusion Models".
A curated collection of papers, explainers, and resources on World Action Models for embodied AI
A Minimal and Elegant Framework & Tutorial for Real-Time Interactive World Models
A Curated List of Awesome Video World Models with AR Diffusion: Covering Algorithms, Applications, and Infrastructure, Aimed at Serving as a Comprehensive Resource for Researchers, Practitioners, a…
[ECCV 2026 Best Paper Award Candidate] LingBot-Map: Geometric Context Transformer for Streaming 3D Reconstruction
Conditional Memory via Scalable Lookup: A New Axis of Sparsity for Large Language Models
The best-benchmarked open-source AI memory system. And it's free.
🏛️ 三省六部制 · OpenClaw Multi-Agent Orchestration System — 9 specialized AI agents with real-time dashboard, model config, and full audit trails
Video PreTraining (VPT): Learning to Act by Watching Unlabeled Online Videos
Official code base for LeWorldModel: Stable End-to-End Joint-Embedding Predictive Architecture from Pixels
Memory Sparse Attention - A scalable, end-to-end trainable latent-memory framework for 100M-token contexts.
[CVPR 2026] SpatialVID: A Large-Scale Video Dataset with Spatial Annotations
[ICML 2026] 4RC: 4D Reconstruction via Conditional Querying Anytime and Anywhere
[ICML 2026] Orienting Latent Actions for Video World Modeling
[CVPR 2025] VideoWorld is a simple generative model that learns purely from unlabeled videos—much like how babies learn by observing their environment.