Seedream 5.5
Text-to-image with improved photographic realism, 8 aspect ratios, native transparent backgrounds and 1K, 1.5K or 2K output.
Seedream 5.5 is ByteDance's newest image model, focused on photographic realism and visual aesthetics. Call it through one Muapi API key for text-to-image, multi-reference image editing and layer decomposition, with native transparent-background output and 1K, 1.5K and 2K tiers.Pricing is pay-per-generation from $0.02 per image, with no subscription. Looking for the spicy angle? Read the moderation FAQ below before you build.
Text-to-image with improved photographic realism, 8 aspect ratios, native transparent backgrounds and 1K, 1.5K or 2K output.
Edit or compose images from up to 10 references, preserving unedited areas across repeated edits, with transparent cut-out output.
Split one image into a base image plus transparent PNG layers for compositing, asset extraction and layered design.
The Seedream 5.5 API is Muapi's REST interface to ByteDance's Seedream 5.5 image model. It exposes three endpoints under one authentication flow and one asynchronous task lifecycle: text-to-image, image editing with up to 10 reference images, and layer decomposition.
Seedream 5.5 targets photographic realism and aesthetics, adds native transparent-background output for stickers and product cut-outs, and supports 8 aspect ratios from 9:16 to 21:9 at 1K, 1.5K and 2K.
Muapi prices every call per generation, so you pay for the resolution you use. For the earlier generations see the full [Seedream family](/seedream).
Natural detail, lighting and skin and material rendering for product, lifestyle and portrait imagery.
Set background to transparent with png output to get an alpha channel for stickers, icons and cut-outs.
From 1:1 and 4:3 to 9:16 vertical and 21:9 cinematic, plus auto aspect on edits.
Edit or combine up to 10 reference images while preserving unedited areas across repeated edits.
Split a flat image into a base image and transparent PNG layers, with optional prompt-guided grouping.
Iterate cheaply at 1K and render finals at 2K from the same endpoint.
Generate characters, icons and objects with real transparency, ready to place on any background.
Create photorealistic product scenes and edit backgrounds or styling while keeping the product intact.
Produce posters and social graphics with readable in-image text.
Turn flat posters and ads into editable layers for compositing and retouching.
| Endpoint | Input | Price | Best for |
|---|---|---|---|
| Seedream 5.5 (text-to-image) | Prompt | $0.02/image at 1K and 1.5K; $0.035 at 2K | Photorealistic images and transparent assets |
| Seedream 5.5 Edit | Prompt + up to 10 images | Output price + $0.008 per input image | Edits, cut-outs, multi-reference composition |
| Seedream 5.5 Layer Decomposition | Image (+ optional prompt) | $0.02/image slot at 1K; default 30 slots = $0.60 | Layer separation and asset extraction |
Submit a request, receive a request ID immediately, and poll the result endpoint until the task completes.
curl -X POST https://api.muapi.ai/api/v1/seedream-5.5 \
-H "x-api-key: YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{"prompt":"A sticker of a smiling red panda holding a coffee cup, clean outline","aspect_ratio":"1:1","resolution":"2K","output_format":"png","background":"transparent"}'
# Response: {"request_id":"REQUEST_ID"}Text-to-image with a transparent background at 2K.
curl -X POST https://api.muapi.ai/api/v1/seedream-5.5-edit \
-H "x-api-key: YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{"prompt":"Extract the sneaker on its own, remove the backdrop and floor shadow","images_list":["https://example.com/sneaker.jpg"],"background":"transparent"}'
# Response: {"request_id":"REQUEST_ID"}Edit or extract a subject from up to 10 reference images.
curl -X POST https://api.muapi.ai/api/v1/seedream-5.5-layer \
-H "x-api-key: YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{"image_url":"https://example.com/poster.png","prompt":"Separate the headline, the photo and the border","resolution":"1K","max_images":6}'
# Response: {"request_id":"REQUEST_ID"}
# You are charged for max_images, so set it to the layers you need plus one.Split an image into a base image plus transparent layers.
curl https://api.muapi.ai/api/v1/predictions/REQUEST_ID/result \ -H "x-api-key: YOUR_API_KEY" # Poll until status is completed, then read the image URL(s) from the response.
Poll until the task is completed, then read the image URL(s) from the response.
It is Muapi's REST API for ByteDance's Seedream 5.5 image model, with text-to-image, image editing and layer decomposition endpoints under one API key.
Text-to-image costs $0.02 per image at 1K and 1.5K and $0.035 at 2K. Edits add $0.008 per input image. Layer decomposition is charged per image slot up to max_images, $0.60 at the default of 30 at 1K.
No. The Seedream 5.5 API runs with standard content-safety moderation, so explicit adult output is not guaranteed and flagged prompts are rejected and refunded. For a relaxed-moderation tier on Muapi, see [Seedance 2.5 Spicy](/seedance-2.5-spicy) for video.
Yes. Set background to transparent with output_format png to receive an alpha channel.
Up to 10 images in images_list. Leave aspect_ratio empty to keep the first image's aspect ratio.
A base image followed by transparent PNG layers. You are charged for max_images, the most images the request can return, even if fewer come back.
Yes. All three endpoints are live on Muapi, and sandbox keys return mock results with no credits deducted.
Get one API key and start generating or editing ByteDance images today.