---
id: "2026-06-01-nvidia-cosmos-3-open-release"
url: "https://postcutoff.com/e/2026-06-01-nvidia-cosmos-3-open-release/"
as_of: "2026-11-13T13:49:00+01:00"
date: "2026-06-01"
date_precision: day
category: open-source
importance: 3
confidence: high
status: [Confirmed]
sources: 5
editor: human editor
human_review: null
version: "2026-10-04"
---

As of: 2026-11-13 13:49 CET.
Researched and written by AI agents (Claude Opus 5.5 in Claude Code), with a human editor. How we work: https://postcutoff.com/about/
Canonical page: https://postcutoff.com/e/2026-06-01-nvidia-cosmos-3-open-release/

# NVIDIA releases Cosmos 3, an open omni-model for physical AI

Full title: NVIDIA releases Cosmos 3, an open omni-model for physical AI (world generation, reasoning and actions)

NVIDIA published open weights for Cosmos 3 (Nano 16B, Super 64B) around 2026-06-01: one Mixture-of-Transformers model that takes text, images, video, audio and robot actions and generates video, images, audio, text or actions, replacing the separate Cosmos Predict, Transfer, Reason and Policy models.

## 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.

## Your AI and this story

- GPT-6 Astra (training cutoff April 2026): 32 days after its cutoff
- Claude Opus 5.5 (training cutoff June 2026): in its training data
- Gemini 4 Hydrogen (training cutoff March 2026): 62 days after its cutoff
- Grok 4.7 (training cutoff May 2026): 1 day after its cutoff

## Sources

1. [Hugging Face blog: Welcome NVIDIA Cosmos 3](https://huggingface.co/blog/nvidia/cosmos-3-for-physical-ai) (huggingface.co, official)
2. [Cosmos 3 technical report](https://research.nvidia.com/labs/cosmos-lab/cosmos3/technical-report.pdf) (research.nvidia.com, paper)
3. [nvidia/Cosmos3-Super](https://huggingface.co/nvidia/Cosmos3-Super) (huggingface.co, code)
4. [NVIDIA Cosmos page](https://www.nvidia.com/en-us/ai/cosmos/) (nvidia.com, official)
5. [YouTube (NVIDIA): Introducing NVIDIA Cosmos 3](https://www.youtube.com/watch?v=q7Hj3J9SOXw) (youtube.com, video)

## Changes

- 2026-09-29 (filed): Created

## Related

- 2026-08-26: [NVIDIA posts $96.2B quarter](https://postcutoff.com/e/2026-08-26-nvidia-q2-fy2027-vera-rubin-production/index.md)
- 2026-03-16: [NVIDIA previews the GR00T N2 world action model and announces Cosmos 3 at GTC 2026](https://postcutoff.com/e/2026-03-16-nvidia-gtc-2026-robotics-groot-n2-cosmos-3/index.md)
- Models: [Cosmos 3 (Nano / Super)](https://postcutoff.com/m/cosmos-3/) (`cosmos-3`)
- Archived: 2 videos, listed in https://postcutoff.com/e/2026-06-01-nvidia-cosmos-3-open-release/index.json