Cosmos 3 Super
- Family
- Cosmos 3
- Released
- 2026-05-31
- Context
- 256K
- Parameters
- 64B
- Architecture
- Mixture of Transformers
- Specialization
- multimodal
- Openness
- Open weights
- License
- OpenMDW 1.1Commercial use: permitted
- Weights
- Available
- Code
- Unknown
- Training
- Pretrained
Cheapest of 1 route · NVIDIA NIM
About
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.
Providers(1)
Compare API pricing across 1 tracked provider route for input and output tokens, batch, and cached reads when available.
| Provider | Input / 1M | Output / 1M | Route |
|---|---|---|---|
| NVIDIA NIM | - | - | ProvisionedPartial |
Available via routers & gateways(1)
Capabilities
Benchmark peer barsfor Long context
No task-mapped benchmark peers are available for this model yet.
Migration checks
No linked migration route is available for this model yet.
API versions
cosmos-3-superCheapest of 1 route · NVIDIA NIM