All Models

NVIDIA logo

NVIDIA Cosmos 3 Super API

NVIDIA Cosmos 3 Super creates images from text and turns reference images into video, with written guidance for subject motion, camera movement, and atmosphere.

Try in Studio
More options (4)
Generate

Pricing depends on the selected inputs. See the rates below.

Supported modes

Animate

Animate uses image and prompt. Adjust the model options to prepare a request for this mode.

video.animate

Text to Image

Text to Image uses prompt. Adjust the model options to prepare a request for this mode.

image.generate

Pricing

Image to video

Price per run for each listed option.

OfferingPrice (USD)
Duration: 1s$0.05 / run
Duration: 2s$0.10 / run
Duration: 3s$0.15 / run
Duration: 4s$0.20 / run
Duration: 5s$0.25 / run
Duration: 6s$0.30 / run
Duration: 7s$0.35 / run
  • Size: 16:9, 4:3, 1:1, 3:4, 9:16 (same price).

Image generation

Price per run for each listed option.

OfferingPrice (USD)
All listed options$0.04 / run
  • Size: 1:1, 3:4, 9:16, 4:3, 16:9 (same price).

Animate options

OptionDetailsDefault
Image *
image
A public URL, a data URI, or an asset id.โ€”
Prompt *
prompt
Text prompt describing the motion and scene of the video to generate.โ€”
Duration
duration
Output video duration in seconds. Options: 1, 2, 3, 4, 5, 6, 7.7
Guidance scale
guidance_scale
Classifier-free guidance scale. Minimum: 0. Maximum: 20.6
Negative prompt
negative_prompt
Content to steer the generation away from.โ€”
Num inference steps
num_inference_steps
Number of denoising steps. Minimum: 1. Maximum: 50.28
Size
size
Output video size preset. Options: 16:9, 4:3, 1:1, 3:4, 9:16."16:9"

NVIDIA Cosmos 3 Super

NVIDIA Cosmos 3 Super combines text-to-image generation with image-to-video animation. A written description defines the subject and visual treatment for a still image. For video, a supplied image establishes the opening frame, and text describes the motion and scene behavior to generate.

Animation guidance can address camera movement as well as action within the scene. That makes the combination useful for developing an image concept and then exploring its movement on screen. A generated still or an existing piece of artwork can serve as the starting frame for this motion study.

What it's good for

Art directors can explore campaign imagery, then animate a selected concept to communicate its intended mood. Filmmakers can test camera ideas around a concept frame. Marketing teams can develop motion treatments for product photographs or character artwork, using the original composition as the basis for a video.

Run NVIDIA Cosmos 3 Super on Protoface

Use NVIDIA Cosmos 3 Super in Protoface Studio without writing code. Choose your inputs and generation settings, then generate with affordable, usage-based pricing. For your own AI app or workflow, use the Protoface API to submit generations and retrieve their outputs.

Model publisher

About NVIDIA

NVIDIA develops AI models, accelerated computing infrastructure, and software for training and deployment. Its work spans Nemotron reasoning models for enterprise applications and the Cosmos world-model family, which supports simulation, video analysis, and synthetic training data for robotics and autonomous vehicle development.

  • World models
  • Physical AI
  • AI infrastructure

Related models

Unified API

One API for generative media.

Protoface serves popular frontier AI models through one API.

  • minimax
  • minimax
  • z.ai
  • z.ai
  • z.ai
  • z.ai
  • z.ai
  • z.ai
  • z.ai

Unified API

One API for generative media.

Protoface serves popular frontier AI models through one API.

  • minimax
  • minimax
  • z.ai
  • z.ai
  • z.ai
  • z.ai
  • z.ai
  • z.ai
  • z.ai

Unified API

One API for generative media.

Protoface serves popular frontier AI models through one API.

  • minimax
  • minimax
  • z.ai
  • z.ai
  • z.ai
  • z.ai
  • z.ai
  • z.ai
  • z.ai