Available nowNew

Seedream 5.5: image editing

By ByteDance

Edit a photo or combine reference images in Seedream 5.5. Describe the specific change and name the details you want to carry into the result.

Model ID
Maker
ByteDance
Added
November 12, 2026

Model id

Use this in every request
bytedance/seedream/v5.5/edit

This model has other modes with their own ids: bytedance/seedream/v5.5/text-to-image.

Run it

Quote the input, start the generation with the quote, then poll the job. Set RAYWAKE_API_KEY first; see Authentication.

# 1. Price the exact input (quotes are valid for up to 10 minutes)
curl -s https://api.raywake.com/v1/quotes \
  -H "Authorization: Bearer $RAYWAKE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "bytedance/seedream/v5.5/edit",
  "input": {
    "prompt": "Replace the grey sofa with a dark green velvet one and add a round wooden side table next to it. Keep the walls, the window and the afternoon light exactly as they are.",
    "image_urls": [
      "https://example.com/input.png"
    ]
  }
}'

# 2. Start the generation with the quote_id from step 1
curl -s https://api.raywake.com/v1/generate \
  -H "Authorization: Bearer $RAYWAKE_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: $(uuidgen)" \
  -d '{"model": "bytedance/seedream/v5.5/edit", "quote_id": "QUOTE_ID", "input": {"prompt":"Replace the grey sofa with a dark green velvet one and add a round wooden side table next to it. Keep the walls, the window and the afternoon light exactly as they are.","image_urls":["https://example.com/input.png"]}}'

# 3. Poll until the status is terminal
curl -s https://api.raywake.com/v1/jobs/JOB_ID \
  -H "Authorization: Bearer $RAYWAKE_API_KEY"

Input

Input fields for bytedance/seedream/v5.5/edit
NameTypeDescription
promptrequiredstring
The text prompt used to edit the image.
image_sizestring (enum)
The size of the image: a preset, or a width and height of 921,600 to 4,624,220 pixels in total, at most 16:1 either way.

Default: auto_2K

One of: square_hd, auto_1K, auto_1.5K, auto_2K

num_imagesinteger
Number of images to generate from the prompt.

Default: 1

output_formatstring (enum)
The file format of the generated image.

Default: png

One of: jpeg, png

watermarkboolean
Whether to add a visible watermark.

Default: false

sync_modeboolean
If `True`, the media will be returned as a data URI and the output data won't be available in the request history.

Default: false

backgroundstring (enum)
Auto lets the model choose from the prompt and the output format. Transparent returns a PNG with an alpha channel and needs PNG output.

Default: auto

One of: auto, opaque, transparent

image_urlsrequiredarray of image URLs
Up to 10 images to edit or combine. Refer to them in the prompt as the first image, the second image and so on.

Output

A succeeded job lists its results under outputs, stored on our side. See Jobs for every field.

Job excerpt
"outputs": [
  { "type": "image", "url": "https://…", "mime_type": "image/png", "thumbnail_url": "https://…" }
]

Pricing

  • 1K or 1.5K · 1 input image$0.07/ image
  • 2K · 1 input image$0.09/ image

The exact charge for your input comes from a quote. How charges settle

API docs

About this model

Overview

Seedream 5.5 turns a prompt into an image at an automatic 1K, 1.5K or 2K size, or at a width and height you set. Ask for a transparent background and it returns a PNG with only the subject on it, ready to place on any design.

Its edit mode takes up to 10 reference images and changes only what you describe, so a room, a product or a face stays as it was around the edit. Each run returns up to six images to choose from.

What you can build

  • Product cut-outs and stickers on a transparent background
  • Room, product and fashion visuals from a written brief
  • Step by step edits of one photo that keep the rest unchanged
  • A product from one photo placed into the scene of another

Prompting tips

  • For a cut-out, choose a transparent background with PNG output and describe only the subject.
  • In an edit, name what to change and what must stay, for example "replace the sofa, keep the walls and the light".
  • With several images, say which one supplies what: the product from the first image, the setting from the second.

Inputs

  • Prompt (required): The text prompt used to edit the image.
  • Image Size (optional): The size of the image: a preset, or a width and height of 921,600 to 4,624,220 pixels in total, at most 16:1 either way. Options: Square HD, Auto 1K, Auto 1.5K, Auto 2K.
  • Number of Images (optional): Number of images to generate from the prompt. From 1 to 6.
  • Output Format (optional): The file format of the generated image. Options: jpeg, png.
  • Background (optional): Auto lets the model choose from the prompt and the output format. Transparent returns a PNG with an alpha channel and needs PNG output. Options: auto, opaque, transparent.
  • Images (required): Up to 10 images to edit or combine. Refer to them in the prompt as the first image, the second image and so on.

Output and limits

  • Image sizes: Square HD, Auto 1K, Auto 1.5K, Auto 2K.
  • File formats: JPEG, PNG.
  • Up to 6 images per run.
  • Small or dense text in the image can come out misspelled.
  • Hands, crowds and complex poses can show artifacts; generate a few variations and pick the best.
  • Some prompts or outputs can be refused by safety checks.

Price guide

  • 1K or 1.5K · 1 input image: $0.07 / image
  • 2K · 1 input image: $0.09 / image