Image generation

prompt must contain 1–5,000 characters. For the Wan 2.7 example, request 1–4 outputs with n; Z-Image Turbo accepts exactly one output per request (n=1). Optionally provide size, watermark, thinking_mode, and region.

Set the image count only with top-level n (default 1). It is the sole quantity used for the quote and the request. If you also send provider parameters, leave out parameters.n; a parameters.n that is not the same integer as n returns HTTP 400 invalid_n before anything is charged.

bash
curl https://api.modelrush.ai/v1/images/generations \
  -H "Authorization: Bearer $MODELRUSH_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"modelrush/wan2.7-image-pro","prompt":"A friendly lunar fox operating an API console","n":1,"size":"1024*1024"}'

Download generated assets into storage you control within 24 hours when durable access is required. ModelRush removes generated outputs and their Provider file references from its database 24 hours after request completion. Keep the ModelRush generation ID and model ID with the asset metadata.

When the task depends on a source image, use Image edits rather than encoding references into a text-only request.

Open a model page for its request parameters and current prices. The four Wan 3 landing pages include a developer API section below the Playground. Read GET /v1/models for current schemas and key-specific access; GET /api/model-pricing returns public prices without authentication; with a valid Bearer API key, it returns the prices applicable to that account.

ModelRush image calls can return HTTP 202 with id and poll_url. Poll that URL with the same API key; GET /v1/predictions/:id completes with status succeeded and outputs. Z-Image Pro Spicy accepts width and height from 256 to 2560 (default 1024), n = 1 and prompt_extend = false by default. Z-Image Spicy also produces one image per request. Do not copy another model's size or output-count limits.