NVIDIA · Robotics
Cosmos 3
Open omnimodel for physical AI in three sizes -- Nano (16B), Super (64B), and a 4B on-device Edge variant -- trained on 20 trillion multimodal tokens with a two-tower Mixture-of-Transformers: an autoregressive tower for vision-language reasoning, a diffusion tower for video/audio generation and robot action prediction. Released under NVIDIA's OpenMDW-1.1 license.
Open the source ↗Similar Agents & Automation releases
- SceneBloom
Yuhan Liu, Lianhui Qin, Shenji Wan · Indoor scene generation · Week 45, 2026
- MotionBricks
NVIDIA (with ETH Zürich, Simon Fraser University and the University of Texas at Austin) · Real-time motion generation · Week 45, 2026
- RoboParty RP1
RoboParty · Open Source Humanoid · Week 43, 2026
- TERRA
EPFL (Mathis Group) · Musculoskeletal Locomotion · Week 42, 2026
- FLUX 3 Action
Black Forest Labs · Robot Action Model · Week 40, 2026
- GPT-Policy
Cheng, Yi, Fang and 12 co-authors · In-Context Robot Learning · Week 39, 2026
More from NVIDIA
- NVIDIA Nemotron 3.5 Transcribe 30B-A4B
NVIDIA · Open multilingual ASR · Week 45, 2026
- SimReady Assets
NVIDIA · Simulation asset dataset · Week 45, 2026
- Nemotron 3.5 Super VL
NVIDIA · Open Multimodal MoE · Week 43, 2026
- SkillSpector v2.11.0
NVIDIA · Agent Skill Scanner · Week 36, 2026
- NVIDIA AVO on ARC-AGI-3
NVIDIA · Agent Orchestration · Week 34, 2026
- NVIDIA Nsight AI
NVIDIA · AI Infrastructure · Week 33, 2026
Published by Quanta AI Labs, an AI agency in India. Browse every week in AI News Weekly.