ByteDance Seedream 5.5
Seedream 5.5 creates images and edits up to 10 references, keeping the subject and untouched areas steady across rounds of edits. Up to 2K, one image per request.
ByteDance Seedream 5.5 (Dola Seedream 5.5) is the newest Seedream image model, generally available on BytePlus ModelArk since 2026-11-12. It generates images from a prompt and edits or blends up to 10 reference images. ByteDance's headline gain over Seedream 5.0 Pro is multi-round editing: across successive edits of one picture the subject, composition and untouched areas stay steady.
Use it for product and campaign imagery, posters and packaging, and chains of edits: pass the previous result as image and say what to change and what to keep.
Inputs: prompt; image and image1 to image9 (up to 10 reference images; with any of them it edits); size 2k (default) or a width x height preset between 0.92 and 4.62 megapixels; watermark (off by default).
Through FlyMyAI it makes one image per request; layer decomposition and the transparent-background switch are not exposed, and it does not make video.
Price: $0.035 per image at 2K or larger, $0.020 at 1536x1536, 1280x720 or 720x1280, plus $0.008 per reference image.
- Text to Image
- Image to Image
Price: From $0.02 per run, depending on size. One run with default settings costs $0.035.
Run ByteDance Seedream 5.5 via API
curl -X POST https://api.flymy.ai/api/v1/flymyai/bytedance-seedream-5_5/predict \
-H "X-API-KEY: $FLYMYAI_API_KEY" \
-H "Content-Type: application/json" \
-d '{"prompt":"A minimalist poster for a jazz night: a golden saxophone on a deep navy background, the headline BLUE NOTE NIGHTS in bold serif letters","size":"1536x1536"}'API documentation · Create a free account · Pricing