infery
← All models

Cosmos 3 Super

cosmos-3-super-text-to-image

Image generationby 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.

images

Details

Accepts
text

Pricing

Price
5 cr / image

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 image every time.
promptrequiredstringText prompt describing the image to generate.
sync_modebooleanIf `True`, the image is returned as a data URI and the output data won't be available in the request history.
image_sizeThe size of the generated image. Each edge is clamped to 512-1280px (multiples of 16).
num_imagesintegerThe number of images to generate.
output_formatstringenum: jpeg, pngThe format of the generated image.
guidance_scalenumberClassifier-free guidance scale. Higher values increase prompt adherence at the cost of diversity.
negative_promptstringContent to steer the generation away from (colors, objects, artifacts).
agentic_early_stopbooleanStop early when a candidate image is already a strong match for the prompt.
num_inference_stepsintegerNumber of denoising steps. More steps yield higher quality but take longer.
enable_safety_checkerbooleanEnable content moderation for the input prompt and generated images. Disabling it requires account authorization; unauthorized requests are always checked, and images flagged as unsafe are returned as black images.
agentic_max_iterationsintegerMaximum number of refinement rounds when agentic generation is enabled.
enable_prompt_expansionboolean
enable_agentic_generationbooleanAutomatically generate and compare multiple candidate images, then refine the prompt between rounds to better match the original request. This can improve prompt adherence but increases latency and billable image generations.
agentic_samples_per_iterationintegerCandidate images to generate and judge per agentic iteration. The best candidate advances to the next rewrite stage.

Output

FieldTypeDescription
seedintegerThe seed used for generation.
imagesarrayThe generated images.
has_nsfw_conceptsarrayWhether each generated image was flagged by the safety checker.