Cosmos-Transfer2.5, built on top of Cosmos-Predict2.5, produces high-quality world simulations conditioned on multiple spatial control inputs.

Cosmos-Predict2.5, the latest version of the Cosmos World Foundation Models (WFMs) family, specialized for simulating and predicting the future state of the world in the form of video.

RLinf is a flexible and scalable open-source infrastructure designed for post-training foundation models (LLMs, VLMs, VLAs) via reinforcement learning.

Code for "ACG: Action Coherence Guidance for Flow-based VLA Models"

[arXiv 2025] Generative View Stitching

ValueCell is a community-driven, multi-agent platform for financial applications.

[ICCV 2023 Oral] ScanNet++: A High-Fidelity Dataset of 3D Indoor Scenes

AgentFlow: In-the-Flow Agentic System Optimization

Code for "Transitive RL: Value Learning via Divide and Conquer"

LITEN: Learning from Inference Time Execution for VLAs

Isaac for Healthcare reference workflows

Public repo for HF blog posts

Official implementation of the paper "GSWorld: Closed-Loop Photo-Realistic Simulation Suite for Robotic Manipulation"

Code implementation of the paper "World-in-World: World Models in a Closed-Loop World"

Ctrl-World: A Controllable Generative World Model for Robot Manipualtion

Official implementation of NeurIPS 2025 paper "SceneWeaver: All-in-One 3D Scene Synthesis with an Extensible and Self-Reflective Agent"

NOF0 - 开源的 AI 交易竞技场

autonomous ai agents for trading in python

The official implementation of Value Flows

Code for DreamControl: Human-inspired Whole-Body Humanoid Control for Scene Interaction via Guided Diffusion

Actor-Free Continuous Control via Structurally Maximizable Q-Functions

An autonomous agent for deep financial research

Contexts Optical Compression

OmniVinci is an omni-modal LLM for joint understanding of vision, audio, and language.