Nvidia Cosmos 3 Super Image-to-Video
Cosmos3 is a collection of Omnimodal world models capable of generating dynamic, high-quality video, image, audio, and action commands from combinations of text, image, video, and action trajectory inputs.
What it does
Nvidia Cosmos 3 Super Image-to-Video, in practice.
NVIDIA Cosmos 3 Super Image-to-Video is NVIDIA's flagship image animation model built on the Cosmos 3 architecture. It transforms a single still image into a smooth, physically grounded video clip guided by a text prompt — delivering cinematic motion, accurate temporal coherence, and fine-grained control over frame count and pacing.
Run Nvidia Cosmos 3 Super Image-to-Video
from $0.083/secParameters
What you can set.
| Parameter | What it does | Options |
|---|---|---|
prompt | Text description of the motion and scene to generate. Maximum 2,500 characters. | |
image_url | The reference image to animate. Accepts a public URL or Base64-encoded image. Supported formats: png, jpeg, jpg, webp. | |
negative_prompt | Text describing content to exclude from the generated video. | |
image_size | Output image resolution. Use a preset size or specify custom width and height via an ImageSize object. | default landscape_16_9 |
duration | The duration of the generated media in seconds (1-7). | 1 2 3 4 5 6 7 |
num_inference_steps | Number of denoising steps. Higher values improve quality but increase generation time. | default 28 |
guidance_scale | Classifier-free guidance scale. Higher values make the output follow the prompt more closely. | default 6.0 |
seed | Random seed for reproducibility. Use the same seed and prompt to reproduce results. | |
enable_safety_checker | Enable or disable the safety checker. Disabling the safety checker may result in NSFW content. | default |
Sample prompt
The prompt behind the sample.
A cute 3D cartoon girl wearing fluffy earmuffs and oversized headphones smiles brightly at the camera. She gently raises her index finger as if sharing an exciting idea, blinking naturally with expressive eyes. Her hair softly sways as a gentle breeze passes by. The camera begins with a medium shot, then slowly pushes into a close-up, capturing her warm smile and sparkling eyes. Soft studio lighting creates delicate shadows while floating dust particles shimmer in the air. The background remains clean and minimalistic with warm brown tones. Pixar-quality animation, cinematic lighting, subtle f
negative_prompt: image_size: landscape_16_9duration: 5num_inference_steps: 28guidance_scale: 6FAQ
Short answers.
How much does Nvidia Cosmos 3 Super Image-to-Video cost?
Pricing starts at $0.083 per second of video at the base resolution; higher resolutions and longer durations cost more. An 8-second clip at the base tier is about $0.66. Usage is billed per request from your balance — no subscription.
Does Nvidia Cosmos 3 Super Image-to-Video run uncensored here?
No. Community applies its own content policy to this model regardless of where it is called from. For a route without an extra platform filter use Uncensored MiniMax H3; this page lists Nvidia Cosmos 3 Super Image-to-Video for everything else it does well.
What does Nvidia Cosmos 3 Super Image-to-Video take as input?
It is a image-to-video model. Cosmos3 is a collection of Omnimodal world models capable of generating dynamic, high-quality video, image, audio, and action commands from combinations of text, image, video, and action trajectory inputs.
How do I use it?
Enter an invite code to open Chat with the model loaded, or join the waitlist. We open seats in batches and track which models are requested most.
Related