Cosmos 3 Super API Reference

Request body

Submit an async generation request. The model field selects the model; other fields are model-specific input parameters.

stringmodelrequired

Model identifier. Set to nvidia/cosmos-3-super/text-to-image.

Default: nvidia/cosmos-3-super/text-to-image

stringpromptrequired

Text prompt describing the image to generate.

Optional<string>aspect_ratio

The aspect ratio of the generated image.

Allowed values: 21:9, 16:9, 3:2, 4:3, 5:4, 1:1, 4:5, 3:4, 2:3, 9:16

Optional<number>guidance_scale

Classifier-free guidance scale. Higher values increase prompt adherence at the cost of diversity.

Range: 0 to 20

Default: 4

Optional<integer>num_inference_steps

Number of denoising steps. More steps yield higher quality but take longer.

Range: 1 to 50

Default: 28

Optional<string>output_format

The format of the generated image.

Allowed values: jpeg, png

Default: jpeg

Optional<integer>seed

The same seed and prompt given to the same model version will produce the same image every time.

Optional<boolean>agentic_early_stop

Stop early when a candidate image is already a strong match for the prompt.

Default: true

Optional<integer>agentic_max_iterations

Maximum number of refinement rounds when agentic generation is enabled.

Range: 1 to 3

Default: 2

Optional<boolean>enable_agentic_generation

Automatically 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.

Default: false

Optional<integer>agentic_samples_per_iteration

Candidate images to generate and judge per agentic iteration. The best candidate advances to the next rewrite stage.

Range: 1 to 3

Default: 2

Optional<integer>num_images

Request parameter supported by this model.

Minimum: 1

Default: 1

Optional<boolean>enable_prompt_expansion

Request parameter supported by this model.

Default: false

Response Schema

The submit endpoint returns a run response. If its status is pending or running, poll GET /v1/run/{id} with the returned opaque ID until it reaches a terminal state.

stringidrequired

Opaque SandBase run identifier. Use it exactly as returned; no prefix is guaranteed.

stringstatusrequired

Current public run status.

Allowed values: pending, running, completed, failed, timeout

Optional<string>model

Public SandBase model name used for this run.

Optional<array<object>>outputs

Present only for completed runs. Each object is capability-specific; inspect the selected model schema for its fields.

Optional<object>error

Present only for failed or timeout runs. Contains a public error type and sanitized message.

Optional<string>error.type

Stable public error category.

Optional<string>error.message

Sanitized error message safe to show to clients.

Optional<object>usage

Usage details when available.

Model capabilities

array<string>capability_tagsrequired

Capabilities declared by the model registry.

Default: text-to-image

stringexecution_moderequired

Execution mode declared by the model registry.

Default: async

# 1. Submit a generation request curl -X POST https://api.sandbase.ai/v1/run \ -H "Authorization: Bearer $SANDBASE_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "nvidia/cosmos-3-super/text-to-image", "prompt": "A photorealistic close-up of two damp hands shaping a spinning cylinder of wet gray clay on a pottery wheel, fingers pinching the walls upward into a narrow-necked vase, water glistening on the clay, soft studio lighting, shallow depth of field.", "aspect_ratio": "21:9", "output_format": "jpeg" }' # 2. If the response is still running, poll the returned id curl https://api.sandbase.ai/v1/run/f3d2e8a1-7c4b-4a12-9d2e-123456789abc \ -H "Authorization: Bearer $SANDBASE_API_KEY"