---
title: Seedream 4.5 | Runware Docs
url: https://runware.ai/docs/models/bytedance-seedream-4-5
description: Seedream 4.5 high fidelity multi reference text to image model
---
# Seedream 4.5

![Seedream 4.5](https://assets.runware.ai/covers/bytedance-seedream-4-5.jpg)

Seedream 4.5 is a ByteDance image model for precise 2K to 4K generation and editing. It improves multi image composition, preserves reference detail, and renders small text more reliably. It supports up to 14 reference images for stable characters and design heavy layouts.

- **Status**: live
- **ID**: `bytedance:seedream@4.5`
- Try in Playground
- **Creator**: ByteDance
- **Release Date**: December 3, 2025
- **Capabilities**: Text to Image, Image to Image, Edit, Checkpoint

## Pricing

$0.04 per run · ~34s per request

**4 example generations**

- [reference-to-image](https://runware.ai/docs/models/bytedance-seedream-4-5/examples#example-submarine-cable-archive-restoration) · 2560×1536: `$0.04` (~44s)
- [text-to-image](https://runware.ai/docs/models/bytedance-seedream-4-5/examples#example-fermentation-culture-lobby-wall-art) · 1536×2560: `$0.04` (~12s)
- [sequential-images](https://runware.ai/docs/models/bytedance-seedream-4-5/examples#example-pinball-factory-archive-colorization) · 4K: `$0.04` (~1m 19s)
- [text-to-image](https://runware.ai/docs/models/bytedance-seedream-4-5/examples#example-mycelium-harvest-supper-invitation) · 1536×2560: `$0.04` (~25s)

times vary, measured end to end · [How pricing works](https://runware.ai/docs/models-api/pricing)

## Compatibility & Validation

Either provide `inputs.referenceImages`, or specify `width/height`.

---

`resolution` cannot be used with `width/height`.

---

When `resolution` is provided, `inputs.referenceImages` is required.

---

`width` and `height` must be used together.

## Notes

**Recommended 2K dimensions:**

| Aspect Ratio | Width × Height |
| --- | --- |
| 1:1 | 2048 × 2048 |
| 4:3 | 2304 × 1728 |
| 3:4 | 1728 × 2304 |
| 16:9 | 2560 × 1440 |
| 9:16 | 1440 × 2560 |
| 3:2 | 2496 × 1664 |
| 2:3 | 1664 × 2496 |
| 21:9 | 3024 × 1296 |

**Recommended 4K dimensions:**

| Aspect Ratio | Width × Height |
| --- | --- |
| 1:1 | 4096 × 4096 |
| 4:3 | 4608 × 3456 |
| 3:4 | 3456 × 4608 |
| 16:9 | 5120 × 2880 |
| 9:16 | 2880 × 5120 |
| 3:2 | 4992 × 3328 |
| 2:3 | 3328 × 4992 |
| 21:9 | 6048 × 2592 |

## Request Parameters

**API Options**

Platform-level options for task execution and delivery.

### [taskType](https://runware.ai/docs/models/bytedance-seedream-4-5#request-tasktype)

- **Type**: `string`
- **Required**: true
- **Value**: `imageInference`

Identifier for the type of task being performed.

### [taskUUID](https://runware.ai/docs/models/bytedance-seedream-4-5#request-taskuuid)

- **Type**: `string`
- **Required**: true
- **Format**: `UUID v4`

UUID v4 identifier for tracking tasks and matching async responses. Must be unique per task.

### [outputType](https://runware.ai/docs/models/bytedance-seedream-4-5#request-outputtype)

- **Type**: `string`
- **Default**: `URL`

Image output type.

**Allowed values**: `URL` `base64Data` `dataURI`

### [outputFormat](https://runware.ai/docs/models/bytedance-seedream-4-5#request-outputformat)

- **Type**: `string`
- **Default**: `JPG`

Specifies the file format of the generated output. The available values depend on the task type and the specific model's capabilities.

- \`JPG\`: Best for photorealistic images with smaller file sizes (no transparency).
- \`PNG\`: Lossless compression, supports high quality and transparency (alpha channel).
- \`WEBP\`: Modern format providing superior compression and transparency support.

> [!NOTE]
> \*\*Transparency\*\*: If you are using features like background removal or LayerDiffuse that require transparency, you must select a format that supports an alpha channel (e.g., \`PNG\`, \`WEBP\`, \`TIFF\`). \`JPG\` does not support transparency.

**Allowed values**: `JPG` `PNG` `WEBP`

### [outputQuality](https://runware.ai/docs/models/bytedance-seedream-4-5#request-outputquality)

- **Type**: `integer`
- **Min**: `20`
- **Max**: `99`
- **Default**: `95`

Compression quality of the output. Higher values preserve quality but increase file size.

### [webhookURL](https://runware.ai/docs/models/bytedance-seedream-4-5#request-webhookurl)

- **Type**: `string`
- **Format**: `uri`

Webhook URL that receives JSON responses via HTTP POST when generation tasks complete. For batch requests with multiple results, each completed item triggers a separate webhook call as it becomes available.

**Learn more** (1 resource):

- [Webhooks](https://runware.ai/docs/models-api/webhooks) (platform)

### [deliveryMethod](https://runware.ai/docs/models/bytedance-seedream-4-5#request-deliverymethod)

- **Type**: `string`
- **Default**: `sync`

Determines how the API delivers task results.

**Allowed values**:

- `sync` Returns results directly in the API response. Over HTTP, a task still running after 120 seconds continues as an async task and comes back with status 'processing'. Poll for its results using getResponse.
- `async` Returns an immediate acknowledgment with the task UUID. Poll for results using getResponse.

**Learn more** (1 resource):

- [Task Polling](https://runware.ai/docs/models-api/task-polling) (platform)

### [uploadEndpoint](https://runware.ai/docs/models/bytedance-seedream-4-5#request-uploadendpoint)

- **Type**: `string`
- **Format**: `uri`

Specifies a URL where the generated content will be automatically uploaded using the HTTP PUT method. The raw binary data of the media file is sent directly as the request body. For secure uploads to cloud storage, use presigned URLs that include temporary authentication credentials.

**Common use cases:**

- **Cloud storage**: Upload directly to S3 buckets, Google Cloud Storage, or Azure Blob Storage using presigned URLs.
- **CDN integration**: Upload to content delivery networks for immediate distribution.

```text
// S3 presigned URL for secure upload
https://your-bucket.s3.amazonaws.com/generated/content.mp4?X-Amz-Signature=abc123&X-Amz-Expires=3600

// Google Cloud Storage presigned URL
https://storage.googleapis.com/your-bucket/content.jpg?X-Goog-Signature=xyz789

// Custom storage endpoint
https://storage.example.com/uploads/generated-image.jpg
```

The content data will be sent as the request body to the specified URL when generation is complete.

### [safety](https://runware.ai/docs/models/bytedance-seedream-4-5#request-safety)

- **Path**: `safety.checkContent`
- **Type**: `object (1 property)`

Content safety checking configuration for image generation.

#### [checkContent](https://runware.ai/docs/models/bytedance-seedream-4-5#request-safety-checkcontent)

- **Path**: `safety.checkContent`
- **Type**: `boolean`

Enable or disable content safety checking. Increases total generation time and adds a small charge on top of the generation cost.

### [ttl](https://runware.ai/docs/models/bytedance-seedream-4-5#request-ttl)

- **Type**: `integer`
- **Min**: `60`

Time-to-live (TTL) in seconds for generated content. Only applies when `outputType` is `URL`.

### [includeCost](https://runware.ai/docs/models/bytedance-seedream-4-5#request-includecost)

- **Type**: `boolean`

Include task cost in the response.

### [numberResults](https://runware.ai/docs/models/bytedance-seedream-4-5#request-numberresults)

- **Type**: `integer`
- **Min**: `1`
- **Max**: `20`
- **Default**: `1`

Number of results to generate. Each result uses a different seed, producing variations of the same parameters.

**Inputs**

Input resources for the task (images, audio, etc). These must be nested inside the \`inputs\` object.

### [referenceImages](https://runware.ai/docs/models/bytedance-seedream-4-5#request-inputs-referenceimages)

- **Path**: `inputs.referenceImages`
- **Type**: `array of strings`
- **Min items**: `1`
- **Max items**: `14`
- **File size (each)**: `up to 10 MB`
- **Size (each)**: `14 px to 6000 px`
- **Aspect ratio (each)**: `1:16 to 16:1`

List of reference images (UUID, URL, Data URI, or Base64).

**Core Parameters**

Primary parameters that define the task output.

### [model](https://runware.ai/docs/models/bytedance-seedream-4-5#request-model)

- **Type**: `string`
- **Required**: true
- **Value**: `bytedance:seedream@4.5`

Send `auto` instead of an AIR identifier to have the model router choose the model for this task. The rest of the request is unchanged, every value you send is kept, and only the parameters you leave out are filled in. See [Model Router](https://runware.ai/docs/models-api/model-router).

**Learn more** (3 resources):

- [Text To Image: Model Selection](https://runware.ai/docs/learn/text-to-image#model-selection) (learn)
- [Image Inpainting: Model Specialized Inpainting Models](https://runware.ai/docs/learn/image-inpainting#model-specialized-inpainting-models) (learn)
- [Image Outpainting: Other Critical Parameters](https://runware.ai/docs/learn/image-outpainting#other-critical-parameters) (learn)

### [positivePrompt](https://runware.ai/docs/models/bytedance-seedream-4-5#request-positiveprompt)

- **Type**: `string`
- **Required**: true
- **Min**: `2`
- **Max**: `3000`

Text prompt describing elements to include in the generated output.

**Learn more** (1 resource):

- [Prompts](https://runware.ai/docs/learn/prompts) (learn)

### [width](https://runware.ai/docs/models/bytedance-seedream-4-5#request-width)

- **Type**: `integer`
- **Required**: conditional — Required when inputs.referenceImages is not provided
- **Min**: `256`
- **Max**: `16383`

Width of the generated media in pixels.

**Learn more** (2 resources):

- [Dimensions](https://runware.ai/docs/learn/dimensions) (learn)
- [Image Outpainting: Dimensions Critical For Outpainting](https://runware.ai/docs/learn/image-outpainting#dimensions-critical-for-outpainting) (learn)

### [height](https://runware.ai/docs/models/bytedance-seedream-4-5#request-height)

- **Type**: `integer`
- **Required**: conditional — Required when inputs.referenceImages is not provided
- **Min**: `256`
- **Max**: `16383`

Height of the generated media in pixels.

**Learn more** (2 resources):

- [Dimensions](https://runware.ai/docs/learn/dimensions) (learn)
- [Image Outpainting: Dimensions Critical For Outpainting](https://runware.ai/docs/learn/image-outpainting#dimensions-critical-for-outpainting) (learn)

### [resolution](https://runware.ai/docs/models/bytedance-seedream-4-5#request-resolution)

- **Type**: `string`

Resolution preset for the output. When used with input media, automatically matches the aspect ratio from the input.

**Allowed values**: `2K` `4K`

### [seed](https://runware.ai/docs/models/bytedance-seedream-4-5#request-seed)

- **Type**: `integer`
- **Min**: `0`
- **Max**: `2147483647`

Random seed for reproducible generation. When not provided, a random seed is generated in the unsigned 32-bit range.

**Settings**

Technical parameters to fine-tune the inference process. These must be nested inside the \`settings\` object.

### [maxSequentialImages](https://runware.ai/docs/models/bytedance-seedream-4-5#request-settings-maxsequentialimages)

- **Path**: `settings.maxSequentialImages`
- **Type**: `integer`
- **Min**: `1`
- **Max**: `15`

Maximum number of sequential images to generate. Combined total of reference images plus sequential images cannot exceed 15. The model may generate fewer images than requested depending on prompt complexity.

**Features**

Standalone addons and post-processing features.

### [watermark](https://runware.ai/docs/models/bytedance-seedream-4-5#request-advancedfeatures-watermark)

- **Path**: `advancedFeatures.watermark`
- **Type**: `object (6 properties)`

Configuration object for adding watermarks to generated videos. Watermarks can be applied using either text or image content with customizable positioning and appearance. You must provide either `text` or `image` content for the watermark, but not both.

**Text watermark**:

```json
"advancedFeatures": {
  "watermark": {
    "text": "© 2025 Company",
    "displayPosition": "bottom-right",
    "opacity": 0.6,
    "fontColor": "#ffffff",
    "bgColor": "#000000"
  }
}
```

**Image watermark**:

```json
"advancedFeatures": {
  "watermark": {
    "image": "c64351d5-4c59-42f7-95e1-eace013eddab",
    "displayPosition": "top-left",
    "opacity": 0.6
  }
}
```

**Tiled watermark**:

```json
"advancedFeatures": {
  "watermark": {
    "text": "PREVIEW",
    "tiled": true,
    "opacity": 0.4,
    "fontColor": "#cccccc"
  }
}
```

#### [text](https://runware.ai/docs/models/bytedance-seedream-4-5#request-advancedfeatures-watermark-text)

- **Path**: `advancedFeatures.watermark.text`
- **Type**: `string`
- **Min**: `2`
- **Max**: `32`

Watermark text.

#### [image](https://runware.ai/docs/models/bytedance-seedream-4-5#request-advancedfeatures-watermark-image)

- **Path**: `advancedFeatures.watermark.image`
- **Type**: `string`

Watermark image (UUID, URL, Data URI, or Base64).

#### [displayPosition](https://runware.ai/docs/models/bytedance-seedream-4-5#request-advancedfeatures-watermark-displayposition)

- **Path**: `advancedFeatures.watermark.displayPosition`
- **Type**: `string`

Watermark position.

**Allowed values**: `top-left` `top-center` `top-right` `center-left` `center-center` `center-right` `bottom-left` `bottom-center` `bottom-right` `tiled`

#### [opacity](https://runware.ai/docs/models/bytedance-seedream-4-5#request-advancedfeatures-watermark-opacity)

- **Path**: `advancedFeatures.watermark.opacity`
- **Type**: `float`
- **Min**: `0.1`
- **Max**: `1`
- **Step**: `0.01`

Watermark opacity.

#### [fontColor](https://runware.ai/docs/models/bytedance-seedream-4-5#request-advancedfeatures-watermark-fontcolor)

- **Path**: `advancedFeatures.watermark.fontColor`
- **Type**: `string`

Text color in hex format.

#### [bgColor](https://runware.ai/docs/models/bytedance-seedream-4-5#request-advancedfeatures-watermark-bgcolor)

- **Path**: `advancedFeatures.watermark.bgColor`
- **Type**: `string`

Background color in hex format.

## Response Parameters

### [taskType](https://runware.ai/docs/models/bytedance-seedream-4-5#response-tasktype)

- **Type**: `string`
- **Required**: true
- **Value**: `imageInference`

Identifier for the type of task this response belongs to.

### [taskUUID](https://runware.ai/docs/models/bytedance-seedream-4-5#response-taskuuid)

- **Type**: `string`
- **Required**: true
- **Format**: `UUID v4`

UUID v4 identifier echoed from the original request, used to match async responses to their tasks.

### [imageUUID](https://runware.ai/docs/models/bytedance-seedream-4-5#response-imageuuid)

- **Type**: `string`
- **Required**: true
- **Format**: `UUID v4`

UUID of the output image.

### [imageURL](https://runware.ai/docs/models/bytedance-seedream-4-5#response-imageurl)

- **Type**: `string`
- **Format**: `uri`

URL of the output image.

### [imageBase64Data](https://runware.ai/docs/models/bytedance-seedream-4-5#response-imagebase64data)

- **Type**: `string`

Base64-encoded image data.

### [imageDataURI](https://runware.ai/docs/models/bytedance-seedream-4-5#response-imagedatauri)

- **Type**: `string`
- **Format**: `uri`

Data URI of the output image.

### [seed](https://runware.ai/docs/models/bytedance-seedream-4-5#response-seed)

- **Type**: `integer`

The seed used for generation. If none was provided, shows the randomly generated seed.

### [NSFWContent](https://runware.ai/docs/models/bytedance-seedream-4-5#response-nsfwcontent)

- **Type**: `boolean`

Flag indicating if NSFW content was detected.

### [cost](https://runware.ai/docs/models/bytedance-seedream-4-5#response-cost)

- **Type**: `float`

Task cost in USD. Present when `includeCost` is set to `true` in the request.

## Examples

### text-to-image (text-to-image)

![Fermentation Culture Lobby Wall Art](https://assets.runware.ai/examples/bytedance-seedream-4-5/317ad63c-4506-40e7-93d2-1096ad58b4cc.jpg)

```json
{
  "taskType": "imageInference",
  "taskUUID": "92fa00ee-e012-4428-8dc2-af08c6c53abe",
  "model": "bytedance:seedream@4.5",
  "positivePrompt": "Create a finished portrait-format fine-art print inspired by a microscopic precision-fermentation culture, designed for the sophisticated lobby of a biotechnology laboratory. An edge-to-edge dark-field composition of hundreds of translucent circular cells, delicate branching membranes, tiny nutrient vesicles, and threadlike microbial structures suspended in deep space. Build a graceful vertical flow: a dense glowing colony near the lower third gradually disperses upward into sparse individual organisms, leaving calm negative space around the upper center. Render the biology with convincing microscopic detail but compose it like abstract contemporary art rather than a textbook diagram. Luminous celadon green, pale saffron, muted coral, and opalescent ivory forms against an inky blue-black ground; subtle halos, diffraction colors, glassy depth, restrained bloom, and crisp cellular edges. Elegant, meditative, scientifically informed, premium gallery aesthetic, richly detailed at close viewing distance, balanced for a tall 3:5 art print. No labels, no typography, no border, no frame, no room mockup, no people, no laboratory equipment.",
  "width": 1536,
  "height": 2560
}
```

---

### text-to-image (text-to-image)

![Mycelium Harvest Supper Invitation](https://assets.runware.ai/examples/bytedance-seedream-4-5/a6d714aa-5026-48f1-9c2c-07414d2f7516.jpg)

```json
{
  "taskType": "imageInference",
  "taskUUID": "f8a683c8-8feb-4c3a-8005-e7db59b05aa8",
  "model": "bytedance:seedream@4.5",
  "positivePrompt": "Create a finished, flat portrait invitation design for an autumn donor dinner hosted by an experimental fermentation research collective. Sophisticated editorial art direction blending nineteenth-century natural-history engraving with contemporary artisan letterpress. Use a warm bone-paper background with visible cotton fibers, deep oxblood typography, saffron accents, muted mushroom brown, and touches of dusty sage. Frame the invitation with an intricate asymmetrical botanical border of oyster mushrooms, barley heads, koji rice grains, black walnuts, tiny fermentation bubbles, ceramic crocks, and curling mycelium threads. The specimens should feel scientifically observed yet abundant and celebratory, rendered with fine engraved lines and selective flat color. Place one large oyster mushroom cluster at the lower right and delicate mycelium filaments climbing toward the upper left, leaving a generous calm central field for typography. Elegant, intimate, cultivated mood; tactile print character; crisp details; balanced negative space; restrained decorative ornament; premium independent cultural-institution aesthetic.\n\nSet the following wording exactly, with no additional text:\n\nTHE MYCELIUM\nHARVEST SUPPER\n\nAn evening of fermentation,\nfieldwork & shared tables\n\nSATURDAY, 17 OCTOBER 2026\nSIX THIRTY IN THE EVENING\n\nThe Fermentation Hall\n48 Foundry Lane\nHudson, New York\n\nDinner begins at seven\nGarden attire\n\nKINDLY REPLY BY 3 OCTOBER\n\nUse a refined high-contrast serif in uppercase for the title, slightly condensed and centered. Set supporting copy in a highly legible classic serif with generous line spacing. Use small caps and subtle saffron rules for the date, venue, and reply line. Maintain impeccable spelling, clean hierarchy, and precise alignment. This is the standalone front of a professionally printed invitation, edge-to-edge artwork, not a card mockup, not shown in a hand, and not placed on a table.",
  "width": 1536,
  "height": 2560
}
```

---

### reference-to-image (reference-to-image)

![Submarine Cable Archive Restoration](https://assets.runware.ai/examples/bytedance-seedream-4-5/c5cd2ffb-d601-4c79-8a60-4aa76371587d.jpg)

```json
{
  "taskType": "imageInference",
  "taskUUID": "728f0fe8-72af-457a-b11d-90d2a158355f",
  "model": "bytedance:seedream@4.5",
  "positivePrompt": "Create a meticulous museum-archive restoration and historically credible colorization. Use reference image 1 as the absolute source for composition, people, facial identity, poses, cable route, beach geography, ship placement, perspective, and documentary character. Repair the torn print, reconstruct the missing lower-right emulsion, remove water stains, creases, dust, scratches, fading, and chemical damage while retaining fine natural film grain and the honest texture of a 1963 press photograph. Use reference image 2 to accurately restore the tracked cable plough: preserve its exact proportions and mechanical construction, with safety-yellow painted bodywork, dark steel tracks, red rollers, winch, hydraulic lines, rivets, wear, and salt-weathered surfaces. Use reference image 3 as the authoritative color reference for the workers, oilskins, knitwear, trousers, gloves, hard hats, marker buoys, skin tones, Atlantic water, and muted Newfoundland coastal palette. Color should feel restrained and photochemically authentic, like a carefully preserved early-1960s Kodachrome print rather than modern digital grading. Recover believable detail in faces without beautifying or changing anyone. Preserve the overcast light, cold wind, wet sand, sea mist, and serious working atmosphere. Do not add, remove, relocate, or redesign any person, machine, cable, vessel, or landscape element. Deliver one seamless restored landscape photograph with natural edge-to-edge framing, no borders, captions, labels, museum display, or added text.",
  "width": 2560,
  "height": 1536,
  "inputs": {
    "referenceImages": [
      "https://assets.runware.ai/assets/inputs/d7482e48-4c22-4c3f-8841-896c845bd0d9.jpg",
      "https://assets.runware.ai/assets/inputs/b7ad7367-2015-44a9-a0f3-1b7f5287ab93.jpg",
      "https://assets.runware.ai/assets/inputs/56717696-09b7-4bb9-abe8-9296a2da860a.jpg"
    ]
  }
}
```

---

### sequential-images (sequential-images)

![Pinball Factory Archive Colorization](https://assets.runware.ai/examples/bytedance-seedream-4-5/8b036ba7-6804-49b6-8045-87447b0d826e.jpg)

```json
{
  "taskType": "imageInference",
  "taskUUID": "22dd2a95-7afa-4a24-9b9d-fa8ab135ac86",
  "model": "bytedance:seedream@4.5",
  "positivePrompt": "Using both references, produce exactly four separate, sequentially generated 4:3 archival photographs—not a collage, grid, contact sheet, restoration comparison, or modern reenactment. Faithfully reconstruct the four exposures shown in the contact sheet while using the enlargement to preserve the technician's exact facial features, kerchief, safety glasses, coveralls, hands, tools, cabinet proportions, machinery, and factory environment. The action must progress clearly: image 1, she aligns and clamps the hinged silk screen over the bare plywood cabinet side; image 2, she pours burnt-orange ink along the upper edge; image 3, she draws the squeegee firmly across the screen; image 4, she lifts the screen and inspects the newly printed geometric motif. Maintain consistent camera height, documentary framing, worker identity, cabinet, print station, background racks, and overhead fluorescent lighting across all four images. Repair tears, creases, scratches, dust, fading, silvering, and missing emulsion with natural photographic detail. Apply restrained, historically plausible 1976 colorization: navy work coveralls, rust-red kerchief, amber safety glasses, warm birch plywood, burnt-orange ink, muted avocado industrial machinery, gray concrete floor, and slightly green fluorescent illumination. Preserve subtle silver-gelatin grain, modest lens softness, realistic skin texture, period materials, and the candid industrial-documentary mood. Do not beautify, modernize, add logos, invent readable signage, or erase all archival texture. Each result must look like an individually restored museum scan from the same original photographic sequence.",
  "resolution": "4K",
  "providerSettings": {
    "bytedance": {
      "maxSequentialImages": 4
    }
  },
  "inputs": {
    "referenceImages": [
      "https://assets.runware.ai/assets/inputs/cf132722-235e-41a2-871c-4dad5f2714f4.jpg",
      "https://assets.runware.ai/assets/inputs/ebd607fa-0b3c-49b9-8cdb-b268e1eb236c.jpg"
    ]
  }
}
```