---
title: FLUX 3 Image Klein Base | Runware Docs
url: https://runware.ai/docs/models/bfl-flux-3-image-klein-base
description: Undistilled open-weights foundation for fine-tuning and customization
---
# FLUX 3 Image Klein Base

![FLUX 3 Image Klein Base](https://assets.runware.ai/covers/bfl-flux-3-image-klein-base.jpg)

FLUX 3 Image Klein Base is the full undistilled checkpoint of Black Forest Labs' open-weights FLUX 3 image model, covering text-to-image generation and reference-based editing. It keeps the complete model capacity, which makes it the variant to fine-tune, customize, and optimize. It suits research, post-training work, and workflows that favor control over generation speed.

- **Status**: live
- **ID**: `bfl:flux@3-image-klein-base`
- Try in Playground
- **Creator**: Black Forest Labs
- **Release Date**: October 15, 2026
- **Zero Data Retention**: Yes
- **Capabilities**: Text to Image, Image to Image, Edit, Checkpoint

## Pricing

$0.0238 from, per run · ~15s per request

**4 example generations**

- [bounding-boxes](https://runware.ai/docs/models/bfl-flux-3-image-klein-base/examples#example-subsea-cable-recruitment-key-art) · 1280×768: `$0.0238` (~9s)
- [text-to-image](https://runware.ai/docs/models/bfl-flux-3-image-klein-base/examples#example-oyster-farm-spectrometer-lifestyle-photo) · 1024×1024: `$0.0238` (~9s)
- [reference-bounding-boxes](https://runware.ai/docs/models/bfl-flux-3-image-klein-base/examples#example-dark-sky-retrofit-social-media-post) · 1024×1280: `$0.0536` (~21s)
- [reference-to-image](https://runware.ai/docs/models/bfl-flux-3-image-klein-base/examples#example-volcanic-observatory-director-portrait) · 1024×1280: `$0.0685` (~28s)

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`.

---

When `inputs.referenceImages` is not provided, `settings.boundingBoxes` is required.

---

When `sourceBox` is provided, `referenceIndex` is required (in `settings.boundingBoxes`).

---

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

## Request Parameters

**API Options**

Platform-level options for task execution and delivery.

### [taskType](https://runware.ai/docs/models/bfl-flux-3-image-klein-base#request-tasktype)

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

Identifier for the type of task being performed.

### [taskUUID](https://runware.ai/docs/models/bfl-flux-3-image-klein-base#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/bfl-flux-3-image-klein-base#request-outputtype)

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

Image output type.

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

### [outputFormat](https://runware.ai/docs/models/bfl-flux-3-image-klein-base#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/bfl-flux-3-image-klein-base#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/bfl-flux-3-image-klein-base#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/bfl-flux-3-image-klein-base#request-deliverymethod)

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

Determines how the API delivers task results.

**Allowed values**:

- `sync` Returns complete results directly in the API response.
- `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/bfl-flux-3-image-klein-base#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/bfl-flux-3-image-klein-base#request-safety)

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

Content safety checking configuration for image generation.

#### [checkContent](https://runware.ai/docs/models/bfl-flux-3-image-klein-base#request-safety-checkcontent)

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

Enable or disable content safety checking. Increases total generation time.

### [ttl](https://runware.ai/docs/models/bfl-flux-3-image-klein-base#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/bfl-flux-3-image-klein-base#request-includecost)

- **Type**: `boolean`

Include task cost in the response.

### [numberResults](https://runware.ai/docs/models/bfl-flux-3-image-klein-base#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.

### [acceleration](https://runware.ai/docs/models/bfl-flux-3-image-klein-base#request-acceleration)

- **Type**: `string`
- **Default**: `medium`

Optimization level.

**Allowed values**: `none` `low` `medium` `high`

### [routing](https://runware.ai/docs/models/bfl-flux-3-image-klein-base#request-routing)

- **Path**: `routing.preferences`
- **Type**: `object (6 properties)`

Model router options. Only read when `model` is `auto`, and never sent to the model.

#### [preferences](https://runware.ai/docs/models/bfl-flux-3-image-klein-base#request-routing-preferences)

- **Path**: `routing.preferences`
- **Type**: `object (3 properties)`

Relative importance of each dimension when models are compared. Weights, not thresholds.

##### [quality](https://runware.ai/docs/models/bfl-flux-3-image-klein-base#request-routing-preferences-quality)

- **Path**: `routing.preferences.quality`
- **Type**: `integer`
- **Min**: `0`
- **Max**: `5`

Importance of one dimension. `0` leaves it out of the decision.

##### [speed](https://runware.ai/docs/models/bfl-flux-3-image-klein-base#request-routing-preferences-speed)

- **Path**: `routing.preferences.speed`
- **Type**: `integer`
- **Min**: `0`
- **Max**: `5`

Importance of one dimension. `0` leaves it out of the decision.

##### [cost](https://runware.ai/docs/models/bfl-flux-3-image-klein-base#request-routing-preferences-cost)

- **Path**: `routing.preferences.cost`
- **Type**: `integer`
- **Min**: `0`
- **Max**: `5`

Importance of one dimension. `0` leaves it out of the decision.

#### [maxPrice](https://runware.ai/docs/models/bfl-flux-3-image-klein-base#request-routing-maxprice)

- **Path**: `routing.maxPrice`
- **Type**: `float`
- **Min**: `0`

Upper bound on what this request may cost, in USD. A model whose cost cannot be bounded is not chosen.

#### [requirements](https://runware.ai/docs/models/bfl-flux-3-image-klein-base#request-routing-requirements)

- **Path**: `routing.requirements`
- **Type**: `string`
- **Max**: `3000`

Conditions the result has to meet, in plain words. A model that cannot meet them is not chosen.

**Inputs**

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

### [referenceImages](https://runware.ai/docs/models/bfl-flux-3-image-klein-base#request-inputs-referenceimages)

- **Path**: `inputs.referenceImages`
- **Type**: `array of strings`
- **Min items**: `1`
- **Max items**: `4`
- **Size (each)**: `16 px or more`

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/bfl-flux-3-image-klein-base#request-model)

- **Type**: `string`
- **Required**: true
- **Value**: `bfl:flux@3-image-klein-base`

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/bfl-flux-3-image-klein-base#request-positiveprompt)

- **Type**: `string`
- **Required**: true
- **Min**: `1`
- **Max**: `10000`

Text prompt describing elements to include in the generated output.

**Learn more** (1 resource):

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

### [negativePrompt](https://runware.ai/docs/models/bfl-flux-3-image-klein-base#request-negativeprompt)

- **Type**: `string`
- **Max**: `10000`

Prompt to guide what to exclude from generation. Ignored when guidance is disabled (CFGScale ≤ 1).

**Learn more** (1 resource):

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

### [width](https://runware.ai/docs/models/bfl-flux-3-image-klein-base#request-width)

- **Type**: `integer`
- **Required**: conditional — Required when inputs.referenceImages is not provided
- **Min**: `128`
- **Max**: `4096`
- **Step**: `16`

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/bfl-flux-3-image-klein-base#request-height)

- **Type**: `integer`
- **Required**: conditional — Required when inputs.referenceImages is not provided
- **Min**: `128`
- **Max**: `4096`
- **Step**: `16`

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)

### [seed](https://runware.ai/docs/models/bfl-flux-3-image-klein-base#request-seed)

- **Type**: `integer`
- **Min**: `0`
- **Max**: `9223372036854776000`

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

**Learn more** (1 resource):

- [Seed](https://runware.ai/docs/learn/seed) (learn)

### [steps](https://runware.ai/docs/models/bfl-flux-3-image-klein-base#request-steps)

- **Type**: `integer`
- **Min**: `1`
- **Max**: `50`
- **Default**: `50`

Total number of denoising steps. Higher values generally produce more detailed results but take longer.

**Learn more** (1 resource):

- [Steps](https://runware.ai/docs/learn/steps) (learn)

### [CFGScale](https://runware.ai/docs/models/bfl-flux-3-image-klein-base#request-cfgscale)

- **Type**: `float`
- **Min**: `1`
- **Max**: `20`
- **Step**: `0.01`
- **Default**: `3.5`

Guidance scale representing how closely the output will resemble the prompt. Higher values produce results more aligned with the prompt.

**Learn more** (1 resource):

- [Cfg Scale](https://runware.ai/docs/learn/cfg-scale) (learn)

**Settings**

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

### [boundingBoxes](https://runware.ai/docs/models/bfl-flux-3-image-klein-base#request-settings-boundingboxes)

- **Path**: `settings.boundingBoxes`
- **Type**: `array of objects (5 properties)`
- **Min items**: `1`

Regions that place, move, resize or remove elements.

#### [description](https://runware.ai/docs/models/bfl-flux-3-image-klein-base#request-settings-boundingboxes-description)

- **Path**: `settings.boundingBoxes.description`
- **Type**: `string`
- **Min**: `1`

What the region holds.

#### [id](https://runware.ai/docs/models/bfl-flux-3-image-klein-base#request-settings-boundingboxes-id)

- **Path**: `settings.boundingBoxes.id`
- **Type**: `string`
- **Required**: true
- **Min**: `1`

Region identifier, unique within the request. Referenced from `positivePrompt` as `<id>`.

#### [referenceIndex](https://runware.ai/docs/models/bfl-flux-3-image-klein-base#request-settings-boundingboxes-referenceindex)

- **Path**: `settings.boundingBoxes.referenceIndex`
- **Type**: `integer | null`
- **Min**: `0`
- **Max**: `9`

Index into `inputs.referenceImages` the element is taken from. Must be less than the number of reference images. Null for a new element.

#### [sourceBox](https://runware.ai/docs/models/bfl-flux-3-image-klein-base#request-settings-boundingboxes-sourcebox)

- **Path**: `settings.boundingBoxes.sourceBox`
- **Type**: `array | null`
- **Min**: `0`
- **Max**: `1000`
- **Min items**: `items: 4`

Position of the element inside the reference as \[top, left, bottom, right\] on a 0 to 1000 scale. Matching `sourceBox` and `targetBox` keep the element in place, different values move or resize it. Null for a new element.

#### [targetBox](https://runware.ai/docs/models/bfl-flux-3-image-klein-base#request-settings-boundingboxes-targetbox)

- **Path**: `settings.boundingBoxes.targetBox`
- **Type**: `array | null`
- **Min**: `0`
- **Max**: `1000`
- **Min items**: `items: 4`

Position in the output as \[top, left, bottom, right\] on a 0 to 1000 scale, whatever the output size. Top must be lower than bottom and left lower than right. Null removes the element, which requires reference images.

**Features**

Standalone addons and post-processing features.

### [watermark](https://runware.ai/docs/models/bfl-flux-3-image-klein-base#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/bfl-flux-3-image-klein-base#request-advancedfeatures-watermark-text)

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

Watermark text.

#### [image](https://runware.ai/docs/models/bfl-flux-3-image-klein-base#request-advancedfeatures-watermark-image)

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

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

#### [displayPosition](https://runware.ai/docs/models/bfl-flux-3-image-klein-base#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/bfl-flux-3-image-klein-base#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/bfl-flux-3-image-klein-base#request-advancedfeatures-watermark-fontcolor)

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

Text color in hex format.

#### [bgColor](https://runware.ai/docs/models/bfl-flux-3-image-klein-base#request-advancedfeatures-watermark-bgcolor)

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

Background color in hex format.

## Response Parameters

### [taskType](https://runware.ai/docs/models/bfl-flux-3-image-klein-base#response-tasktype)

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

Identifier for the type of task this response belongs to.

### [taskUUID](https://runware.ai/docs/models/bfl-flux-3-image-klein-base#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/bfl-flux-3-image-klein-base#response-imageuuid)

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

UUID of the output image.

### [imageURL](https://runware.ai/docs/models/bfl-flux-3-image-klein-base#response-imageurl)

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

URL of the output image.

### [imageBase64Data](https://runware.ai/docs/models/bfl-flux-3-image-klein-base#response-imagebase64data)

- **Type**: `string`

Base64-encoded image data.

### [imageDataURI](https://runware.ai/docs/models/bfl-flux-3-image-klein-base#response-imagedatauri)

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

Data URI of the output image.

### [seed](https://runware.ai/docs/models/bfl-flux-3-image-klein-base#response-seed)

- **Type**: `integer`

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

### [NSFWContent](https://runware.ai/docs/models/bfl-flux-3-image-klein-base#response-nsfwcontent)

- **Type**: `boolean`

Flag indicating if NSFW content was detected.

### [cost](https://runware.ai/docs/models/bfl-flux-3-image-klein-base#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)

![Oyster Farm Spectrometer Lifestyle Photo](https://assets.runware.ai/examples/bfl-flux-3-image-klein-base/24b4ac30-ffaa-49ef-80f1-83a892480c20.jpg)

```json
{
  "taskType": "imageInference",
  "taskUUID": "8e8da38a-44ff-4734-b341-f26fef6a76b0",
  "model": "bfl:flux@3-image-klein-base",
  "positivePrompt": "High-end commercial product lifestyle photograph for a marine field-instrument campaign. On a working oyster farm at low tide, a shellfish farmer in teal waterproof bib overalls and textured orange rubber gloves uses a compact handheld water-quality spectrometer. Make the unbranded instrument the clear hero: a rugged matte ivory rectangular body with rounded corners, dark sensor window, one cobalt-blue control dial, small recessed display showing only simple colored measurement bars, and a black wrist tether. The farmer holds the device prominently in the lower-center foreground just above a shallow stainless sampling cup filled with seawater; anatomically correct hands and a natural working grip. Behind it, orderly oyster baskets recede across reflective tidal flats, with a small aluminum work skiff and another farmer softly out of focus. Cold blue dawn atmosphere cut by a low band of coral sunrise, wet surfaces sparkling, faint sea mist, authentic mud and salt residue, practical documentary realism elevated with precise advertising lighting. Square composition, close environmental portrait, eye-level 50mm perspective, shallow depth of field, crisp product materials and controls, believable scale, refined teal, ivory, rust-orange and silver palette, calm purposeful mood, generous breathing room around the device. Finished standalone photograph, no typography, no logo, no border, no mockup.",
  "negativePrompt": "extra fingers, malformed hands, duplicated device, floating object, oversized product, unreadable labels, brand marks, prominent text, futuristic holograms, sterile studio background, fantasy scenery, excessive blur, plastic-looking water, oversaturated colors, illustration, CGI render",
  "width": 1024,
  "height": 1024
}
```

---

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

![Volcanic Observatory Director Portrait](https://assets.runware.ai/examples/bfl-flux-3-image-klein-base/a5944ac7-125b-4e2a-aa9d-0eb7660beef0.jpg)

```json
{
  "taskType": "imageInference",
  "taskUUID": "fc3516c7-d1fd-4e9f-a481-cfbfe8929ed4",
  "model": "bfl:flux@3-image-klein-base",
  "positivePrompt": "Create a finished vertical professional editorial portrait using both references. Preserve the exact facial identity, age, natural skin texture, short salt-and-pepper hairstyle, eye shape, and distinctive smile lines of the woman in reference image 1. Portray her as the director of a national volcanic observatory, photographed for an institutional annual report and international press profile. Place her within the observatory environment and adopt the sophisticated slate-blue, graphite, and warm-amber visual language of reference image 2.\n\nFrame her from mid-torso upward in a confident three-quarter stance, shoulders relaxed, body turned slightly away while her eyes meet the camera. Her expression is calm, incisive, and quietly reassuring rather than smiling broadly. Dress her in a beautifully tailored deep-charcoal collarless technical blazer over a matte rust-colored crew-neck top; understated and credible for a senior field scientist, with no badge, logo, jewelry, or safety costume. Keep the clothing realistic and impeccably fitted.\n\nUse the distant volcanic ridge through the window as subtle environmental context, softly out of focus and positioned behind one shoulder without appearing dramatic or dangerous. Include only restrained hints of the control room: a blurred aluminum instrument cabinet and a small amber practical light. No readable monitors or interface graphics. Light her face with a large soft key light from camera left, gentle cool window fill, and a faint warm edge light motivated by the room. Maintain dimensional but flattering facial modeling, accurate catchlights, realistic pores, fine hair detail, and natural color.\n\nPremium institutional portrait photography, contemporary science editorial art direction, 85mm portrait-lens character, shallow depth of field, crisp focus on both eyes, subtle analog tonal roll-off, muted slate blue and graphite background with one rust accent, composed negative space, dignified and authoritative mood. The result must look like a commissioned photograph, not a composite or illustration. No lettering, captions, logos, borders, UI, or watermark.",
  "negativePrompt": "different person, altered facial structure, younger appearance, beauty retouching, plastic skin, exaggerated smile, glamour makeup, fashion pose, lab coat, hard hat, respirator, disaster scene, erupting lava, smoke emergency, science-fiction control room, neon lighting, readable screens, text, logo, watermark, extra people, distorted hands, asymmetrical eyes, soft eyes, oversharpening, HDR",
  "width": 1024,
  "height": 1280,
  "inputs": {
    "referenceImages": [
      "https://assets.runware.ai/assets/inputs/94e880d1-cc07-4a4a-abe5-534b850b3dbe.jpg",
      "https://assets.runware.ai/assets/inputs/4d1e1515-5d65-4da2-a596-0eb5c0a8d9b0.jpg"
    ]
  }
}
```

---

### bounding-boxes (bounding-boxes)

![Subsea Cable Recruitment Key Art](https://assets.runware.ai/examples/bfl-flux-3-image-klein-base/c12d5015-cc2f-4e34-818f-08acab1f3acf.jpg)

```json
{
  "taskType": "imageInference",
  "taskUUID": "abc826c4-b5f2-4a34-bf91-62ffd429bd9e",
  "model": "bfl:flux@3-image-klein-base",
  "positivePrompt": "Finished landscape anime key art for a global telecommunications recruitment campaign. The <subject> is a full-body female subsea cable technician standing securely on the right side of the composition, wearing a practical white atmospheric diving suit with safety-orange shoulder panels, articulated joints, utility harness, reinforced boots, and a clear domed helmet; her calm face is visible through the visor, one hand holds a compact diagnostic scanner, and the other steadies a thick fiber-optic cable. Keep her entire silhouette inside the designated region and make her the unmistakable main character. Surround her with a clearly organized deep-ocean maintenance site: a dark cable trench runs diagonally from the lower left toward her boots, bright cyan inspection nodes punctuate the cable, a small unmanned repair submersible hovers in the open water at left, and a rectangular service gantry behind the technician visually frames her controlled placement. Distant hydrothermal rock formations and tiny drifting particles establish immense depth without clutter. Premium contemporary anime illustration, crisp confident linework, believable industrial engineering, cel-shaded forms with softly painted environmental depth, cobalt and midnight-blue ocean, electric-cyan work lights, restrained safety-orange accents, dramatic lateral beams, subtle bioluminescence, focused and quietly heroic mood. Wide cinematic composition with generous negative space on the left, strong depth layers, polished campaign-ready finish, no lettering, no logos, no interface, no border.",
  "negativePrompt": "photorealism, chibi proportions, fantasy armor, exposed skin, casual clothing, extra people, duplicate character, cropped body, malformed hands, tangled unreadable machinery, oversized weapons, monsters, text, typography, logos, watermark, app interface, comic panels",
  "width": 1280,
  "height": 768,
  "settings": {
    "boundingBoxes": [
      {
        "id": "subject",
        "description": "main subject described in the prompt",
        "targetBox": [
          150,
          550,
          850,
          950
        ]
      }
    ]
  }
}
```

---

### reference-bounding-boxes (reference-bounding-boxes)

![Dark-Sky Retrofit Social Media Post](https://assets.runware.ai/examples/bfl-flux-3-image-klein-base/3d77d190-dab4-4331-a07d-537a870235f6.jpg)

```json
{
  "taskType": "imageInference",
  "taskUUID": "c032d001-f866-4a08-92f5-1e88c6ccb110",
  "model": "bfl:flux@3-image-klein-base",
  "positivePrompt": "Create a finished 4:5 social media campaign image for a municipal dark-sky streetlight retrofit program. Move <element> into the right side of the composition while preserving the owl's recognizable face, pose, proportions, cream-and-copper feather pattern, feet, and photographic identity from the reference. Place the owl naturally on the curved arm of a modern shielded streetlight in the right-side target region. Build a bold editorial photo-composite around it: an evening neighborhood seen from a slightly low angle, deep indigo sky, subtle silhouettes of apartment roofs and leafy street trees, and a restrained field of small stars. The shielded fixture casts a precise warm amber pool downward onto the pavement without spilling light into the sky. On the left, show a faint hazy glow from distant unshielded lighting fading into the cleaner dark sky toward the owl, creating an immediate visual comparison without a literal split screen. Sophisticated public-interest campaign art direction, cinematic realism combined with crisp screen-printed grain, limited palette of midnight blue, warm amber, chalk cream, and muted coral, strong graphic shapes, confident negative space, accessible and hopeful mood. Keep the owl as the sharp visual hero and integrate it convincingly with matching rim light and contact shadow. Standalone image with no words, logos, signage, borders, interface elements, or buttons.",
  "negativePrompt": "extra owls, duplicate animal, altered facial disk, deformed wings, missing feet, floating owl, oversized streetlight, text, letters, logo, watermark, app interface, photorealistic daytime sky, fantasy elements, neon rainbow palette, cluttered background",
  "width": 1024,
  "height": 1280,
  "settings": {
    "boundingBoxes": [
      {
        "id": "element",
        "description": "primary subject from reference image 0",
        "referenceIndex": 0,
        "sourceBox": [
          150,
          100,
          850,
          500
        ],
        "targetBox": [
          150,
          500,
          850,
          900
        ]
      }
    ]
  },
  "inputs": {
    "referenceImages": [
      "https://assets.runware.ai/assets/inputs/112380fa-5ea4-4b30-b3dd-c7fe77ca436a.jpg"
    ]
  }
}
```