Mixer AI
Seedream 5.5
Create and edit images at 1K, 1.5K and 2K with up to 10 input photos and eight aspect ratios.
When to choose it
Seedream 5.5 in Mixer AI creates images from a text description and edits uploaded photos. It offers 1K, 1.5K and 2K sizes, eight aspect ratios and up to ten input photos in JPEG, PNG or WEBP, each up to 10 MB. The model is available in the Telegram bot, Studio, Canvas and through MCP.
- The same request works as creating a new image and as editing from photos.
- Up to 10 references and eight aspect ratios, including 21:9.
Modes and settings
- From text alone, or editing with 1–10 input photos.
- 1K, 1.5K and 2K; ratios 1:1, 4:3, 3:4, 16:9, 9:16, 2:3, 3:2 and 21:9.
Price in coins
| Quality / size | Input | Coins / generation | Extra image, coins / image |
|---|---|---|---|
| 1K | Photo editing | — | — |
| 1K | Text | — | — |
| 1.5K | Photo editing | — | — |
| 1.5K | Text | — | — |
| 2K | Photo editing | — | — |
| 2K | Text | — | — |
Coins are Mixer AI balance units. The table shows a base request without extra variations. Multiple results and new attempts can increase the total; check it before submitting.
When editing, every input image is billed separately at the rate in the extra-image column. The surcharge is added before rounding the complete total. Resolution refers to the output.
The dollar price follows the current coin rate.
What you can make
- A scene that combines a person, a product and a setting from several photos.
- Editing a finished frame: change the background, clothing or one detail and keep the rest.
How to direct the result
- Give each photo a role: identity, clothing, object, background or composition.
- For edits, list what must stay unchanged first, then describe one change.
Example: assemble a product shot from three photos
Create a vertical product shot. Use the bottle shape from image one, the label color from image two, and the pale stone surface from image three. Keep the bottle proportions and logo placement. Soft side light, realistic reflections, no extra objects.
What to keep in mind
- This model has no 4K size: the maximum is 2K.
- References do not guarantee exact text, logos or fine details, so check them in the result.
Sources
Guides and comparisons using this model
FAQ
How many photos can I upload?
Up to ten input images in the editing mode.
Which file formats and sizes are supported?
JPEG, PNG and WEBP, each file up to 10 MB.
Which result sizes are available?
1K, 1.5K and 2K.
Can I start without a photo?
Yes. A text description is enough to create an image from scratch.
What prompt length is allowed?
From 5 to 5,000 characters.
Where can I use the model?
In the Telegram bot, Studio, Canvas and through MCP.
What changes the price?
It depends on version, quality and mode. The table prices one generation at the listed settings. Additional results, new attempts and chargeable input materials can increase the total.
Do I need a separate model subscription?
No. In Mixer AI, you top up one coin balance and use it for the tasks you need. You do not need a separate subscription to this model.