Using Cosmos 3 Super on NVIDIA NIM
Implementation guide · Cosmos 3 · NVIDIA AI
NVIDIA NIM exposes Cosmos 3 Super through model ID cosmos3-reasoner-super. Use the setup steps, sourced pricing, capabilities, and official provider links below to validate this route before deployment.
Last refreshed 2026-06-01. Next refresh: weekly.
Quick Start
- 1
- 2Use the NVIDIA NIM SDK or REST API to call
cosmos3-reasoner-super— see the documentation for request format.
Code Examples
Pricing on NVIDIA NIM
| Type | Price (per 1M) |
|---|---|
| Image input | $1.00 |
| Video input | $1.00 |
| Audio input | $1.00 |
Capabilities
About Cosmos 3 Super
Cosmos 3 Super is NVIDIA's flagship 64B-parameter omnimodel for physical AI, designed for large-scale synthetic data generation and high-fidelity simulation on NVIDIA Hopper and Blackwell datacenter GPUs. Architecture: dual-tower Mixture-of-Transformers with a 32B autoregressive Reasoner and a 32B diffusion-based Generator. Supports 256K token reasoning context, 720p video generation at variable frame rates, and 10+ robot embodiment action domains. Ranked #1 among open models on Physics-IQ, PAI-Bench, R-Bench, RoboLab, RoboArena, VANTAGE-Bench, TAR, and Artificial Analysis image/video leaderboards (Computex 2026). Training data: 1.3B data points across 393 datasets (2024-2026). Inference performance (vLLM-Omni): ~55s for 50-step video on 8xH200.