ByteDance

Seedream 5

Seedream 5 accepts a text prompt for new images and optional image references for editing or style guidance. Pro configurations expose 1K and 2K output, while Lite is oriented around 2K and 3K output and includes separate generation and editing entries in this catalog. Up to ten references may be used, but fewer well-chosen images usually make each reference's purpose easier to communicate.

How to use this model

  1. 1

    Choose Pro or Lite and confirm that the selected entry matches generation or editing.

  2. 2

    Write the desired scene or edit as a precise prompt with subject, composition, lighting, palette, and required text.

  3. 3

    For editing, add only relevant reference images and explain what each one contributes.

  4. 4

    Select an aspect ratio, supported size, and output format before generation.

  5. 5

    Check text, subject identity, and small structural details, then refine the instruction instead of stacking contradictory prompts.

5.0 LiteAvailable

5.0 Lite - Image Editing

Seedream 5 Lite Edit applies a natural-language instruction to reference images and returns an image, not a video. It is intended for high-quality 2K or 3K edits with up to ten reference images. Keep the prompt focused on the requested transformation and describe what must be preserved.

Try now

Properties

  • Image-editing workflow; output is an image, not video.
  • Supports 2K or 3K output and up to ten references.

How to use this model

  1. 1

    Upload the source and only the references needed for the edit.

  2. 2

    Describe the change and list elements that must stay unchanged.

  3. 3

    Choose aspect ratio, 2K or 3K size, and output format.

  4. 4

    Use one focused edit per iteration when fidelity is important.

  5. 5

    Compare identity, layout, text, and boundaries with the source at full size.

Examples

A vintage tram on a quiet Lisbon street
Urban scene produced with a Seedream editing workflow.

Parameters

Prompt

prompt
Required

The text prompt describing the desired image.

Number of Images

num_images
Optional

Number of separate model generations to be run with the prompt. Default value: 1

Default: 1Range: 1 – 6

Maximum Number of Images

max_images
Optional

If set to a number greater than one, enables multi-image generation. The model will potentially return up to max_images images every generation, and in total, num_images generations will be carried out. In total, the number of images generated will be between num_images and max_images*num_images. Default value: 1

Default: 1Range: 1 – 6

Image URLs

image_urls
Required

List of URLs of input images for editing. Presently, up to 10 image inputs are allowed.

Image Size

image_size
Optional
Default: auto_2K

  • square_hd

  • square

  • portrait_4_3

  • portrait_16_9

  • landscape_4_3

  • landscape_16_9

  • auto_2K

  • auto_3K

Enable Safety Checker

enable_safety_checker
Optional
Default: true

Sync Mode

sync_mode
Optional
Default: false

Return source URLs

return_byteplus_urls
Optional
Default: false
5.0 LiteAvailable

5.0 Lite - Text to Image

Despite the legacy identifier, this catalog entry is Seedream 5 Lite text-to-image generation and returns an image, not video. It supports 2K or 3K output, common aspect ratios, and standard image formats. Use the separate Lite Edit entry when an existing image must be modified.

Try now

Properties

  • Text-to-image workflow; output is an image, not video.
  • Supports 2K or 3K output.

How to use this model

  1. 1

    Write a complete text-to-image prompt with subject, composition, lighting, palette, and style.

  2. 2

    Specify exact visible wording when the scene includes text.

  3. 3

    Choose aspect ratio, 2K or 3K size, and output format.

  4. 4

    Generate alternatives to compare composition before adding minor detail.

  5. 5

    Switch to the Lite Edit entry if source-image preservation is required.

Examples

A mechanical butterfly on a glowing orchid
Macro fantasy detail generated with Seedream.

Parameters

Prompt

prompt
Required

The text prompt describing the desired image.

Number of Images

num_images
Optional

Number of separate model generations to be run with the prompt. Default value: 1

Default: 1Range: 1 – 6

Maximum Number of Images

max_images
Optional

If set to a number greater than one, enables multi-image generation. The model will potentially return up to max_images images every generation, and in total, num_images generations will be carried out. In total, the number of images generated will be between num_images and max_images*num_images. Default value: 1

Default: 1Range: 1 – 6

Image Size

image_size
Optional
Default: auto_2K

  • square_hd

  • square

  • portrait_4_3

  • portrait_16_9

  • landscape_4_3

  • landscape_16_9

  • auto_2K

  • auto_3K

Enable Safety Checker

enable_safety_checker
Optional
Default: true

Sync Mode

sync_mode
Optional
Default: false
Seedream 5 ProAvailable

Seedream 5 Pro

Seedream 5 Pro creates images from prompts and can use optional reference images for editing or style guidance. It supports 1K and 2K output, common aspect ratios, and JPEG, PNG, or WebP. Up to ten references may be supplied, though a focused set usually makes the intended transformation clearer.

Try now

Properties

  • Supports text-to-image and reference-guided editing.
  • Supports 1K and 2K output and up to ten reference images.

Best for

  • High-quality text-to-image generation
  • Image editing with reference images

Tips

  • Use 2K when the additional resolution justifies the higher price.
  • The first reference image is included; every additional reference image increases generation cost.

How to use this model

  1. 1

    Describe the target image or edit with precise composition and text requirements.

  2. 2

    Add references only when they have a clear subject, style, or edit role.

  3. 3

    Select aspect ratio, 1K or 2K size, and output format.

  4. 4

    For editing, say explicitly what must remain unchanged.

  5. 5

    Review text, identity, and small details before refining the prompt.

Examples

Example output from Seedream 5 Pro
A reviewed image output created with Seedream 5 Pro in Omnipix.

Parameters

Prompt

prompt
Required

Reference Images

image_urls
Optional

Aspect Ratio

aspect_ratio
Optional
Default: 1:1

  • 1:1

  • 16:9

  • 9:16

  • 4:3

  • 3:4

  • 3:2

  • 2:3

  • 4:5

  • 5:4

  • 21:9

Output Format

output_format
Optional
Default: jpeg

  • jpeg

  • png

  • webp

Size

size
Optional
Default: 1K

  • 1K

  • 2K

Seedream 5 Pro LRAvailable

Seedream 5 Pro LR

LR means Less Restrictive. This version keeps Seedream 5 Pro generation, reference editing, 1K/2K output, aspect-ratio, and format controls while applying substantially more permissive prompt filtering, including prompts up to NSFW content. Platform rules still apply.

LR = Less Restrictive: prompts up to NSFW level are supported, subject to platform rules.
Try now

Properties

  • Supports generation and reference editing with Pro output controls.
  • Less-restrictive filtering, including prompts up to NSFW level.

Best for

  • High-quality text-to-image generation
  • Image editing with reference images

Tips

  • Use 2K when the additional resolution justifies the higher price.
  • The first reference image is included; every additional reference image increases generation cost.

How to use this model

  1. 1

    Use the same precise scene or edit brief as for the standard Pro entry.

  2. 2

    Add only relevant references and explain their roles.

  3. 3

    Choose aspect ratio, 1K or 2K size, and output format.

  4. 4

    Use LR deliberately when the request needs less-restrictive filtering, including NSFW-level content.

  5. 5

    Inspect text, identity, and unintended reference changes.

Parameters

Prompt

prompt
Required

Reference Images

image_urls
Optional

Aspect Ratio

aspect_ratio
Optional
Default: 1:1

  • 1:1

  • 16:9

  • 9:16

  • 4:3

  • 3:4

  • 3:2

  • 2:3

  • 4:5

  • 5:4

  • 21:9

Output Format

output_format
Optional
Default: jpeg

  • jpeg

  • png

  • webp

Size

size
Optional
Default: 1K

  • 1K

  • 2K