Images / NVIDIA

Cosmos3 Super · Text2Image

A large open-weight image generator aimed at physical-AI scenes, with structured prompt support and publisher-tested GB200 or multi-H100 deployment.

Verified sourceReleased May 31, 2026Source checked 10/14/2026
LOCALRENTED GPUOPEN WEIGHTS
Freshness 59 out of 100. Release age: 137 days old, May 31, 2026. Measures age, not quality.
Popularity 53 out of 100. 6,305 downloads in 30 days; 195 lifetime likes. Checked Oct 15, 2026. Measures adoption, not quality.
Arena Open Source Image generation: rank 4 of 19. Ranked model: Cosmos3-Super-Text2Image. Export dated Oct 7, 2026. Source checked Oct 15, 2026. Human preference ranking, not hardware suitability.
Category rankings & sources

Each position belongs to a separate Arena category. These are human preference rankings—not hardware recommendations. We do not combine the categories into one score.

Image generation#4of 19 open models
Cosmos3-Super-Text2ImagenvidiaLicense: OpenMDW-1.1
Export dated Oct 7, 2026Checked: Oct 15, 2026Overall board: #33
View official Image generation board

Rankings reflect the dated export; live Arena boards can change. Open-source positions use Arena’s open-model filter. The overall position also includes the other models on that category board. Availability of weights and license terms still need to be checked separately.

What do these scores mean?

Two ways to compare models. Neither score measures answer quality, benchmark performance, or whether a model fits your machine.

Freshness · release age

A new release starts at 100. The score halves every 180 days: 100 × 2^(−age in days / 180). When the release date is unavailable, a repository date is a fallback—not proof of when the model was released. Missing dates stay unscored.

Check the date source
Popularity · adoption

Downloads over the past 30 days carry 80% of the score; lifetime likes carry 20%. Both use a logarithmic scale, capped at 10 million downloads and 100,000 likes. If likes are unavailable, downloads carry all the weight. Missing download statistics stay unscored.

round(80 × min(1, log10(1 + downloads) / 7) + 20 × min(1, log10(1 + likes) / 5))

Check the statistics source
Scores as of Oct 15, 2026. Statistics checked Oct 15, 2026. Statistics are marked for refresh after 72 hours.Full scoring method
WORKFLOWS FOR THIS MODEL 1

What do you want to do?

Choose a task. Get the software, files, settings, and steps to run it.

MAKE IT PERSONALChoose your hardware

Open a guide for its software, hardware requirements, and setup steps.

At a glance

Parameters
64B
Architecture
Mixture of Transformers
License
OpenMDW 1.1
Software
Diffusers, vLLM-Omni
MY HARDWARE

Will it run on your machine?

Save your machine to see a personalized rating and its reasoning.

Add my hardware

Best for

  • Structured image generation for physical-AI scene ideation.

Tradeoffs

  • Outputs are not physically accurate simulations or safety-certified data.
  • A substantial multi-GPU or GB200 setup makes this a specialist route rather than the default image recommendation.

Ways to run it

Diffusers Cosmos3OmniPipeline · Linux · GB200 server

Publisher says its Diffusers example was tested on GB200.

Official instructions

vLLM-Omni · Linux · 8× H100 server

Publisher documents an 8×H100 inference endpoint. Do not present this as a single consumer GPU setup.

Official instructions

Explore its uses

IMAGE GENERATIONPHYSICAL AISTRUCTURED PROMPTSRENTED GPU
REFERENCE LIBRARY

Sources and files

Primary sources and files behind this model page.

1 SOURCE

Keep exploring

Cosmos3 Super · Text2Image: hardware, VRAM & setup | YouRunAI