Qwen-Image-2.1
- Pricing
- Not published
- Released
- 5mo ago
Cost
Cosmos3-Super-Text2Image is NVIDIA's post-trained text-to-image variant of its Cosmos3-Super base model, part of the Cosmos 3 collection of omnimodal world foundation models. NVIDIA positions Cosmos 3 as an omnimodal world-model platform for Physical AI and says this variant generates high-fidelity images that match a given text description. It is intended for Physical AI applications such as robotics, autonomous driving, and smart spaces, including industrial and factory-scale settings.
How it compares
Frequently asked questions
Quality and benchmarks
The closest alternatives, matched on model kind and release date with no comparable price, are Qwen-Image-2.1 by Alibaba, Muse Image by Meta, and Qwen Image 2.1 Turbo by Qwen.
Qwen-Image-2.1Cosmos3-Super-Text2Image vs Qwen-Image-2.1Muse ImageCosmos3-Super-Text2Image vs Muse ImageQwen Image 2.1 TurboCosmos3-Super-Text2Image vs Qwen Image 2.1 Turbo
Licensing and access
Cosmos3-Super-Text2Image is open weights, released under the OpenMDW-1.1 license.
You can download the Cosmos3-Super-Text2Image weights from the Hugging Face repository nvidia/Cosmos3-Super-Text2Image.
Basics
Cosmos3-Super-Text2Image was made by NVIDIA and released on May 31, 2026.
Yes, Cosmos3-Super-Text2Image is open weights, so you can download and run the model yourself.
Media
No, image editing with an image input is not listed for Cosmos3-Super-Text2Image.
On the Arena Text to Image leaderboard, Cosmos3-Super-Text2Image holds rank 36 with a score of 1150.
Generated from Vertical Tabs data, updated November 11, 2026.
Discussion
No comments yet
Be the first to share your thoughts.