NVIDIA releases Cosmos 3, an open omni-model for physical AI
Confirmed
Status
- Claim
Confirmed
- Our reporting
- High confidence
- Importance
- 3 of 5
- Last verified
- 29 September 2026
Your AI and this story
- GPT-6 Astra32 days after its cutoff
- Claude Opus 5.5In its training data
- Gemini 4 Hydrogen62 days after its cutoff
- Grok 4.71 day after its cutoff
It is after the cutoff of GPT-6 Astra, Gemini 4 Hydrogen and Grok 4.7, and can be in the training data of Claude Opus 5.5.
Key facts
- Sizes: Cosmos3-Nano 16B, Cosmos3-Super 64B; Hugging Face, license OpenMDW-1.1 (commercial use allowed), no gating
- Inputs: text, images, video, audio, action trajectories; outputs: text, images, video (5-400 frames), 48 kHz audio, actions
- Architecture: Mixture-of-Transformers combining autoregressive and diffusion transformers
- NVIDIA: best open text-to-image and image-to-video model on Artificial Analysis and best policy model on RoboArena
- Announced at GTC 2026-03-16; HF repos went public 2026-05-31; technical report dated 2026-06-22
What happened
Cosmos 3 is NVIDIA’s first single world foundation model that can generate worlds, reason about physics and output actions. Before it, developers had to chain separate models: Predict 2.5, Transfer 2.5, Reason 2 and Policy.
Why it matters
It is the largest open world model aimed at robotics and autonomous vehicles. It also shows the field moving toward models that do both “world simulation” and “policy” in one, a direction GR00T N2 is also expected to follow.
Sources
5 sources from 4 sites. Numbers match the chips in the text.
5 sources: 4 primary, 1 reaction
Primary
- Hugging Face blog: Welcome NVIDIA Cosmos 3huggingface.co, official
- Cosmos 3 technical reportresearch.nvidia.com, paper
- nvidia/Cosmos3-Superhuggingface.co, code
- NVIDIA Cosmos pagenvidia.com, official
Reactions
- YouTube (NVIDIA): Introducing NVIDIA Cosmos 3youtube.com, video
Changes
- Filed
