Seedream 5.0 Image API: Pro, Flash, and pricing

TTAPI added two dated Seedream model IDs: doubao-seedream-5-0-pro-260628 and doubao-seedream-5-0-flash-260915. They are separate Seedream variants on the TTAPI model page. The exact IDs are the values to send in the model field.

The official Seedream 5.0 Pro page highlights dense information graphics, interactive editing, photographic texture, multilingual generation, and broad creative styles. The official gallery on the TTAPI detail page keeps Pro and Flash examples labeled by their source model so comparisons remain readable.

The asynchronous POST /sd/api/v3/images/generations/tasks route accepts model and prompt. Add image for one reference image or image-to-image editing, and pass two to ten images for multi-image fusion. The same family also exposes documented layer-decomposition and interactive-edit workflows, where the prompt describes layers or annotated regions.

The synchronous POST /sd/api/v3/images/generations route returns images directly. Generation supports 1K, 1.5K, 2K (default), or explicit pixel dimensions. Layer decomposition instead defaults to auto, accepts 1K, 1.5K, or 2K, and does not accept pixel dimensions; its layers are always PNG. optimize_prompt_options.mode defaults to standard, and fast is not supported by Flash. Async workflows additionally list execution_expires_after, priority, and safety_identifier; no callback field is published.

The async request below uses TTAPI's English OpenAPI definition. The workflow schema returns a task id. Save it for polling, and keep the TT-API-KEY header on your server.

Retrieve the task with GET /sd/api/v3/images/generations/tasks/{id}, replacing {id} with the submitted task identifier. The documented statuses are PENDING_QUEUE, ON_QUEUE, SUCCESS, and FAILED. On success, data.images contains the image URLs. response_format defaults to url, with download links valid for 24 hours; b64_json requests Base64 data. Bound polling and handle failures or expired jobs.

curl --request POST \
  --url 'https://api.ttapi.io/sd/api/v3/images/generations/tasks' \
  --header "TT-API-KEY: $TTAPI_KEY" \
  --header 'Content-Type: application/json' \
  --data '{
  "model": "doubao-seedream-5-0-pro-260628",
  "prompt": "A high-density infographic explaining the six major types of tea, with a watercolor flavor wheel and labeled brewing temperatures.",
  "size": "2K",
  "output_format": "jpeg",
  "response_format": "url",
  "watermark": false
}'

TTAPI currently lists Seedream Pro with a limited-time 70% discount. Pro generation is 1.35 quota at 1.5K and 2.7 quota at 2K; layer split is 0.675 quota at 1.5K and 1.35 quota at 2K. Input images are listed at 0.09 quota. Seedream Flash is listed at 1.8 quota per image.

The promotion is resolution and workflow specific. Confirm the live pricing page before estimating a batch, especially when mixing normal generation, layer decomposition, and image inputs. The examples on this page are official Seedream showcases, not a benchmark of TTAPI generation quality or cost.

curl --request POST \
  --url 'https://api.ttapi.io/sd/api/v3/images/generations' \
  --header "TT-API-KEY: $TTAPI_KEY" \
  --header 'Content-Type: application/json' \
  --data '{
  "model": "doubao-seedream-5-0-pro-260628",
  "prompt": "A clean editorial poster about the six major types of tea, with readable labels and a watercolor flavor wheel.",
  "size": "2K",
  "output_format": "jpeg",
  "response_format": "url",
  "background": "opaque",
  "watermark": false
}'