ByteDance: Seedream 4.0
seedream-v4- Price
- From $0.03
- Modalities
- Image
About Seedream 4.0
Seedream 4.0 is ByteDance Seed's image model that merges text-to-image generation and general-purpose editing in one architecture. It accepts text, images or any combination, takes up to about a dozen reference images, produces multi-image sequences and renders text including charts and formulas. ByteDance raised maximum output from 2K to 4K and reports a more than tenfold speedup over Seedream 3.0.
Where it works well
- Unified generation and editing, which ByteDance presents as the central design change from Seedream 3.0.
- Multi-image reference fusion with up to a dozen reference pictures to carry characters, styles and structures.
- Multi-image output sequences for sets of related pictures from one prompt.
- Text rendering, including formulas and charts, listed among the eight capabilities in the launch post.
- Adaptive aspect ratios and 4K output according to ByteDance.
When to choose another model
- It predates Seedream 4.5 and 5.0, which ByteDance continues to develop.
- Visual reasoning and online search arrive with the 5.0 Lite release, not in this version.
- Output size follows the aspect ratio of your prompt or input image rather than fixed size presets.
Getting started
Create API key
Create a key in Console, then use it with every model on the platform.
Send your first request
Copy the example for your language and run it against the endpoint.
Image to imageText to imageTokenLab endpointPOST/v1/images/generationscurl -X POST "https://api.tokenlab.sh/v1/images/generations" \ -H "Authorization: Bearer sk-xxx" \ -H "Content-Type: application/json" \ -d '{ "operation": "text-to-image", "model": "seedream-v4", "prompt": "A minimalist product photo of matte black headphones on a soft blue background." }'
Parameters
Image to image
| Parameter | Values and limits | Default | Notes |
|---|---|---|---|
aspect_ratio | Not provided | Not provided | Model-specific aspect-ratio values and the default are not yet verified for this public operation. |
async | Not provided | Not provided | Not provided |
image | Not provided | Not provided | This public operation accepts source images. The maximum total source-image count is not yet verified. |
image_url | Not provided | Not provided | Not provided |
n | Not provided | 1 | Defaults to one output image. The model-specific maximum is not declared in this public contract. |
operation | Not provided | Not provided | Not provided |
prompt | Not provided | Not provided | Not provided |
resolution | Not provided | Not provided | Model-specific resolution values and the default are not yet verified for this public operation. |
response_format | Not provided | Not provided | The public image response selector accepts url or b64_json. A URL-only supply rejects b64_json before upstream execution. Omit this field to keep the selected supply response representation; no uniform omitted-field default is declared here. |
size | Not provided | Not provided | Model-specific size values and the default are not yet verified for this public operation. |
Use /v1/images/edits for image-to-image.
Text to image
| Parameter | Values and limits | Default | Notes |
|---|---|---|---|
aspect_ratio | Not provided | Not provided | Model-specific aspect-ratio values and the default are not yet verified for this public operation. |
async | Not provided | Not provided | Not provided |
n | Not provided | 1 | Defaults to one output image. The model-specific maximum is not declared in this public contract. |
operation | Not provided | Not provided | Not provided |
prompt | Not provided | Not provided | Not provided |
resolution | Not provided | Not provided | Model-specific resolution values and the default are not yet verified for this public operation. |
response_format | Not provided | Not provided | The public image response selector accepts url or b64_json. A URL-only supply rejects b64_json before upstream execution. Omit this field to keep the selected supply response representation; no uniform omitted-field default is declared here. |
size | Not provided | Not provided | Model-specific size values and the default are not yet verified for this public operation. |
style | Not provided | Not provided | Not provided |
Use /v1/images/generations for text-to-image.
Use /v1/images/generations for text-to-image and /v1/images/edits for image-to-image.
Pricing
The Verified price applies to Verified, which costs less on most models. The Official price is the model maker's published price and applies to the more reliable Official route. Auto bills the route that completes the request.
Text to image / Image to image
per image- Official price
- $0.03
- Verified price
- —
- Discount
- —
| Official priceper image | Verified priceper image | Discount | |
|---|---|---|---|
| Text to image / Image to image | $0.03 | — | — |
Usage & activity
Success rate is the share of requests that completed. Latency is how long a full response takes; P95 means 95% of requests finished within that time.
Usage & availability
Last 24 hours- Requests
- Success rate
- P95 latency
- Total tokens
Data is based on aggregate user requests, excluding status checks.
Open in Console
Open Seedream 4.0 in Console with a prompt ready to edit or send.
Help me create an image with seedream-v4 using text-to-image at /v1/images/generations. Show the result, cost, and any limits I should know.
Use cases
Best for- Text to Image
Multi-reference composites
Combine a character sheet, a style reference and a location photo into one coherent scene.
Storyboard image sets
Generate a series of connected frames in a single request for a comic or presentation.
Chart and formula visuals
Make teaching images, such as a chalkboard derivation or a labelled chart, where equations and labels need to read correctly.
Prompt examples
Use the two reference photos: put the woman from image 1 on the bridge from image 2 at sunset.
Generate four frames of a fox crossing a snowy road, same fox in each, wide to close.
A blackboard with the quadratic formula written clearly and a parabola sketched beside it.
FAQ
What is Seedream V4?
Seedream V4 is Seedream 4.0, ByteDance Seed's unified image generation and editing model. It accepts text, images or both as input, so one model covers new pictures and edits to existing ones.
How many reference images does it take?
ByteDance's launch post says it accepts up to a dozen reference images at a time, enough to carry several characters, styles and locations into one picture.
Can it output several images at once?
ByteDance lists multi-image output sequences among its capabilities, so one prompt can yield several related pictures, such as frames of a story with a consistent subject.
What resolution does Seedream 4.0 reach?
ByteDance states the maximum output rose from 2K to 4K, with adaptive aspect ratios that follow the input or the prompt rather than a fixed canvas.
Should I move to 4.5 or 5.0?
Compare on your own prompts. Seedream 4.5 refines editing consistency, and the 5.0 models add reasoning, search and design focus. Seedream 4.0 remains fine for simple generation and edits.
How much does Seedream 4.0 cost?
On TokenLab, Seedream 4.0 costs $0.03 per image. The pricing table above shows the full breakdown. Rates depend on the billing unit, specification, and usage. Compare matching conditions in the model's detailed pricing; a single rate does not determine the total cost.
Which endpoint should Seedream 4.0 use?
Use https://api.tokenlab.sh/v1/images/generations for Seedream 4.0. The request example below shows the matching code shape.
Which operations does Seedream 4.0 support?
Seedream 4.0 supports Image to image, Text to image. Select an operation above to see its endpoint and request example.
Compare Seedream 4.0
Sources
Reviewed Oct 2, 2026