Seedream 5.0 Lite: Image-to-Image

ByteDanceimagelive~47s

Transform images intelligently with detailed text prompts.

avg$0.035/ run

An estimate, trued up after the run. Failed runs cost $0.00.

How this model is billed

Quoted before you run. A failed run costs $0.00.

Typical run ~47s: the measured mean, end to end.

Compare every price

Run Seedream 5.0 Lite: Image-to-Image

Seedream 5.0 Lite: Image-to-Image API

Async: the POST answers at once with a request id. Poll it, or take the webhook.

Docs

Submit

Parameters at the top level of the JSON body. Answers 202 with a request id.

curlPOST /v1/models/seedream-v5-lite-image-to-image

curl -X POST https://api.ezflows.io/v1/models/seedream-v5-lite-image-to-image \
  -H "Authorization: Bearer $EZFLOWS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "a wide shot of a desert at dusk",
    "aspect_ratio": "16:9",
    "size": "3K"
  }'

# 202 Accepted, at once — it never waits for the model:
# {"request_id": "req_8f2a41c9b70e4d15", "status": "queued",
#  "poll_url": "https://api.ezflows.io/v1/requests/req_8f2a41c9b70e4d15"}

Poll

Same bearer header, every few seconds, until the status is final.

curlGET /v1/requests/:id

REQUEST_ID=req_8f2a41c9b70e4d15   # from the POST

while :; do
  BODY=$(curl -s https://api.ezflows.io/v1/requests/$REQUEST_ID \
    -H "Authorization: Bearer $EZFLOWS_API_KEY")
  STATUS=$(echo "$BODY" | jq -r .data.status)
  case "$STATUS" in succeeded|failed|cancelled) break ;; esac
  sleep 3
done

echo "$BODY" | jq -r '.data.output.artifacts[].url'   # media.ezflows.io, kept 90 days

Response

Files are on media.ezflows.io, kept 90 days. A failed run says why and costs nothing.

json200 · succeeded

{
  "data": {
    "id": "req_8f2a41c9b70e4d15",
    "model_slug": "seedream-v5-lite-image-to-image",
    "status": "succeeded",
    "http_status": 200,
    "cost": 0.035023,
    "auth_method": "api_key",
    "error": null,
    "submitted_at": "2026-09-12T18:07:21.114Z",
    "queue_started_at": "2026-09-12T18:07:21.402Z",
    "execution_started_at": "2026-09-12T18:07:22.010Z",
    "first_byte_at": "2026-09-12T18:07:52.871Z",
    "finished_at": "2026-09-12T18:07:55.203Z",
    "io_deleted_at": null,
    "canvas_id": null,
    "canvas_node_id": null,
    "client_attempt_id": null,
    "idempotency_key": null,
    "input": {
      "prompt": "a wide shot of a desert at dusk",
      "aspect_ratio": "16:9",
      "size": "3K"
    },
    "output": {
      "artifacts": [
        {
          "index": 0,
          "kind": "image",
          "contentType": "image/png",
          "url": "https://media.ezflows.io/outputs/<account>/req_8f2a41c9b70e4d15/0.png"
        }
      ]
    }
  }
}

Webhook

Skip polling: we POST the result to your endpoint, at least once.

Register: Create an endpoint under Webhooks in the dashboard; the signing secret is shown once.
Events: request.succeeded · request.failed · request.cancelled
Headers: ezflows-event-id (dedupe on it) · ezflows-timestamp (unix seconds) · ezflows-signature t=<ts>,v1=<hex HMAC-SHA256 of "<ts>.<raw body>">
Verify: Over the exact bytes received, in constant time; reject anything older than five minutes.
Respond: Any 2xx within 10 s. Redirects are not followed.
Retries: Five, with backoff: 30 s · 60 s · 2 min · 4 min · 8 min. Disabled after five consecutive dead deliveries.

jsonDelivery body

{
  "id": "82cf46a4-841e-41d6-b3c7-facc937d3252",
  "type": "request.succeeded",
  "created_at": "2026-09-12T18:07:55Z",
  "data": {
    "id": "req_8f2a41c9b70e4d15",
    "model_slug": "seedream-v5-lite-image-to-image",
    "status": "succeeded",
    "cost": 0.035023,
    "output": {
      "artifacts": [
        {
          "url": "https://media.ezflows.io/…/0.png"
        }
      ]
    }
  }
}

curlverify.sh

# Create the endpoint under Webhooks in the dashboard; the signing secret is shown once.
# ezflows-signature: t=<ts>,v1=<hex HMAC-SHA256 of "<ts>.<raw body>">
# body.json = the exact bytes received. Reject anything older than five minutes.
EXPECTED=$({ printf '%s.' "$TS"; cat body.json; } \
  | openssl dgst -sha256 -hmac "$EZFLOWS_WEBHOOK_SECRET" -r | cut -d' ' -f1)

[ "$EXPECTED" = "$V1" ] && echo "valid" || echo "reject"

Parameters 6

Name	Type	Default	Allowed	Description
promptrequired	string	—	—	Describe the desired image transformation or style change.
image_input	string[]	—	image urls	Input images to guide creation, Upto 10 Images for reference.
aspect_ratio	enum	16:9	1:1 · 4:3 · 3:4 · 16:9 · 9:16 · 3:2 · 2:3 · 21:9	Image aspect ratio. 16:9 for widescreen, 1:1 for square.
size	enum	3K	2K · 3K	Image resolution. 2K for moderate detail, 3K for fine detail.
max_images	integer	1	—	Number of images to generate. Use 1 for single output.
optimize_prompt	enum	fast	standard · fast	Select optimization mode: 'fast' for quicker results, 'standard' for detailed transformation.

File inputs (image_input: image urls) take URLs. Upload the raw file to POST /v1/uploads with its own Content-Type and pass the url it returns. Inputs are kept 90 days.

About this model

## Overview Seedream 5.0 Lite Image-to-Image, from ByteDance, transforms images intelligently from detailed text prompts. Supply up to 10 input images as references and describe the result you want: a restyle, a variation, a new setting, or a composition that combines several sources. ## What it does well - **Reference-driven generation.** Up to 10 input images guide the result. - **Detailed transformations.** Describe the change in the prompt and the model applies it to your input. - **Two resolutions.** 2K for moderate detail and 3K, the default, for fine detail. - **Eight aspect ratios,** from square to 21:9 and 9:16. - **Prompt optimisation.** Fast mode for quicker results, standard for more detailed transformation. - **A watermark toggle,** off by default. ## Prompting tips Say what should change and what should stay: "keep the composition and turn the photo into a warm watercolour illustration". When you pass several images, explain what each contributes. Use fast optimisation to explore, and standard when you want the most detailed transformation. ## Limits to know The output follows the aspect ratio you choose, and the 2K or 3K setting sets how much detail it carries, so pick the ratio of the place the image will be used. This model is billed on the real cost of each run, and the price is quoted on the Run button before you spend. For generating from text alone, use the Seedream 5.0 Lite text-to-image model. ## FAQ ### How many input images can I use? Up to 10. ### What is the difference between fast and standard optimisation? Fast returns results more quickly; standard works harder on detailed transformations.