Cosmos3-H-Dreams

nvidia/cosmos3-h-dreams

Action-conditioned surgical world model distilled from Cosmos3-H-Embodied-Nano. It takes a surgical scene image and instrument action vectors and generates streaming video at 832×480 and 10 fps. The released student uses four denoising steps per latent frame.

Compare

Context window: -
Input: Image · Other
Output: Video
Released: 2026-10-20

Weights: Open
License: NVIDIA Open Model License

Pricing

No official pricing available.

Estimate usage cost

Compare verified public prices for this exact model and region. Enter uncached input and cache usage separately. Estimates exclude unspecified fees, credits and currency conversion. Set requests to zero to estimate token charges only.

Input tokens

Output tokens

Cache read tokens

Cache write tokens

Requests

Context tokens (optional)

Currency

Batch processing

Performance

No evaluation results available.

Providers

No providers found

ModelRegistry · Catalog updated 2026-10-21