infery
← All models

Cosmos 3 Super Image to Video

cosmos-3-super-image-to-video

Video editby NVIDIA

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.

Example

Details

Accepts
image

Pricing

Price
6.25 cr / second

Prices in credits (1 credit = $0.01).

Data schema

Input

FieldTypeDescription
seedThe same seed and prompt given to the same model version will produce the same video every time.
promptrequiredstringText prompt describing the motion and scene of the video to generate.
image_urlrequiredstringURL of the conditioning first-frame image for the video.
sync_modebooleanIf `True`, the video is returned as a data URI and the output data won't be available in the request history.
image_sizeThe size of the generated video. The request is clamped and snapped to the nearest supported NVIDIA tier (256p/480p/720p) and aspect ratio.
num_framesintegerNumber of frames to generate. More frames yield a longer video.
guidance_scalenumberClassifier-free guidance scale. Higher values increase prompt adherence at the cost of diversity.
negative_promptstringContent to steer the generation away from (artifacts, unwanted motion). Defaults to NVIDIA's recommended i2v negative prompt; pass an empty string to disable.
frames_per_secondintegerFrames per second of the output video.
agentic_early_stopbooleanStop the agentic loop early when the critic score clears the strict quality threshold.
num_inference_stepsintegerNumber of denoising steps. More steps yield higher quality but take longer.
enable_safety_checkerbooleanEnable content moderation for the input prompt and image. Disabling it requires account authorization; unauthorized requests are always checked.
agentic_max_iterationsintegerMaximum agentic prompt stages when agentic generation is enabled.
enable_prompt_expansionbooleanIf true, the Cosmos3-Nano Reasoner (a VLM that sees the first frame) rewrites the prompt into the dense caption Cosmos3 was trained on. The app starts a local Reasoner by default, or uses COSMOS_PROMPT_UPSAMPLER_BASE_URL when configured. Falls back to the raw prompt if expansion fails.
enable_agentic_generationbooleanEnable the iterative Cosmos agentic loop: prompt upsampling, candidate video generation, VLM critique of sampled frames, and prompt rewrite. Each candidate is a full render, so this is substantially slower and costlier than a single generation.
agentic_samples_per_iterationintegerCandidate videos to generate and judge per agentic iteration. The best candidate advances to the next rewrite stage.

Output

FieldTypeDescription
seedintegerThe seed used for generation.
videoThe generated video.