---
title: GPT-Image-2.6 Sunburst | Runware Docs
url: https://runware.ai/docs/models/openai-gpt-image-2-6-sunburst
description: Image generation and editing for detailed compositions and precise edits
---
# GPT-Image-2.6 Sunburst

![GPT-Image-2.6 Sunburst](https://assets.runware.ai/covers/openai-gpt-image-2-6-sunburst.jpg)

GPT-Image-2.6 Sunburst is OpenAI's image model for detailed compositions and editing work that needs precise control. Besides raster images, it generates SVG graphics and animated GIFs from text, and it edits images from marks drawn over them, like a circle around the object to change. A thinking level sets how much it plans before rendering, separate from rendering quality.

- **Status**: live
- **ID**: `openai:gpt-image@2.6-sunburst`
- Try in Playground
- **Creator**: OpenAI
- **Release Date**: November 5, 2026
- **Capabilities**: Text to Image, Image to Image, Edit, Vectorize, Checkpoint, Transparency

## Pricing

~$0.0703 typical run

**Rates** (per 1M tokens)

- Text input: `$5.00`
- Image input: `$8.00`
- Image output: `$30.00`
- SVG authoring · GPT-5.4 Mini · Input: `$0.75`
- SVG authoring · GPT-5.4 Mini · Output: `$4.50`
- SVG authoring · GPT-5.4 · Input: `$2.50`
- SVG authoring · GPT-5.4 · Output: `$15.00`
- SVG authoring · GPT-5.5 · Input: `$5.00`
- SVG authoring · GPT-5.5 · Output: `$30.00`
- SVG authoring · GPT-5.6 Terra · Input: `$2.00`
- SVG authoring · GPT-5.6 Terra · Output: `$12.00`
- SVG authoring · GPT-6 Luna · Input: `$0.10`
- SVG authoring · GPT-6 Luna · Output: `$0.50`
- SVG authoring · GPT-6 Astra · Input: `$10.00`
- SVG authoring · GPT-6 Astra · Output: `$50.00`

**6 example generations**

- [text-to-gif](https://runware.ai/docs/models/openai-gpt-image-2-6-sunburst/examples#example-geothermal-plant-director-looping-portrait): `$0.0374` (~31s)
- [text-to-image](https://runware.ai/docs/models/openai-gpt-image-2-6-sunburst/examples#example-subsea-cable-landing-wall-art): `$0.0431` (~1m 1s)
- [text-to-vector](https://runware.ai/docs/models/openai-gpt-image-2-6-sunburst/examples#example-nocturnal-moth-survey-festival-poster): `$0.0634` (~1m)
- [inpaint](https://runware.ai/docs/models/openai-gpt-image-2-6-sunburst/examples#example-cave-rescue-equipment-sticker-pack): `$0.0771` (~57s)
- [transparent-background](https://runware.ai/docs/models/openai-gpt-image-2-6-sunburst/examples#example-oyster-farm-childrens-comic-panel): `$0.0788` (~32s)
- [reference-to-image](https://runware.ai/docs/models/openai-gpt-image-2-6-sunburst/examples#example-adaptive-rainwear-fashion-lookbook): `$0.192` (~1m 51s)

billed per token, so a request costs what the model produced · times vary, measured end to end · [How pricing works](https://runware.ai/docs/models-api/pricing)

## Compatibility & Validation

Inside `inputs`, `maskImage` cannot be used with `markupImage`.

---

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

---

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

---

When `outputFormat` is `SVG, or GIF`, `inputs.referenceImages` cannot be used.

---

When `outputFormat` is `GIF`, `width` must be at most `960` and `height` must be at most `960`.

---

When `settings.background` is `transparent`, `outputFormat` is limited to `PNG`, `WEBP`.

---

When `settings.gif` is provided, `outputFormat` must be `GIF`.

---

When `settings.svg` is provided, `outputFormat` must be `SVG`.

---

When `settings.svg.model` is `openai:gpt@5.4-mini, openai:gpt@5.4, or openai:gpt@5.5`, `settings.svg.thinkingLevel` is limited to `none`, `low`, `medium`, `high`, `xhigh`.

---

When `settings.svg.model` is `openai:gpt@5.6-terra, or openai:gpt@6-luna`, `settings.svg.thinkingLevel` is limited to `none`, `low`, `medium`, `high`, `xhigh`, `max`.

---

When `settings.svg.model` is `openai:gpt@6-astra`, `settings.svg.thinkingLevel` is limited to `low`, `medium`, `high`, `xhigh`, `max`.

---

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

## Notes

GPT-Image-2.6 Sunburst picks its mode from the inputs you attach and the output format you ask for:

- **Text-to-image:** Provide a prompt on its own.
- **Editing:** Provide `inputs.referenceImages`. The first image is the one being edited, and any others bring in what it does not have.
- **Targeted editing:** Add `inputs.maskImage` or `inputs.markupImage` to an edit. Both point at part of the first reference image.
- **Vector and animation:** Set `outputFormat` to `SVG` or `GIF`, from a prompt alone.

**Two thinking levels:** `settings.thinkingLevel` sets how much the image model plans the composition, separately from the rendering effort in `settings.quality`. An SVG request carries a second, independent level in `settings.svg.thinkingLevel`, the reasoning of the language model that writes the SVG, and the values it accepts depend on `settings.svg.model`.

**SVG output:** the image model first renders a reference image, which `settings.quality` and `settings.thinkingLevel` shape, and the author model writes the SVG from it. The author's tokens are billed at that model's own rate on top of the image tokens. Without `width` and `height`, the model picks the canvas.

**GIF output:** every frame is drawn on one sheet and cut out of it, which is why the largest frame size shrinks as `settings.gif.frameCount` grows. Without `settings.gif`, a GIF has 9 frames shown for 0.2 seconds each, and without `width` and `height` the frames are 512 × 512. The file loops forever.

**Markup:** `inputs.markupImage` is a transparent PNG with marks drawn over the first reference image, such as a circle around the object to change. The prompt refers to the mark, as in "the lamp inside the red circle", and asks for it to be removed from the result.

## Request Parameters

**API Options**

Platform-level options for task execution and delivery.

### [taskType](https://runware.ai/docs/models/openai-gpt-image-2-6-sunburst#request-tasktype)

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

Identifier for the type of task being performed.

### [taskUUID](https://runware.ai/docs/models/openai-gpt-image-2-6-sunburst#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/openai-gpt-image-2-6-sunburst#request-outputtype)

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

Image output type.

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

### [outputFormat](https://runware.ai/docs/models/openai-gpt-image-2-6-sunburst#request-outputformat)

- **Type**: `string`
- **Required**: conditional — Required when settings.background is provided or settings.gif is provided or settings.svg is provided
- **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.
- \`GIF\`: Animated image format (no audio), suitable for short loops or previews.
- \`SVG\`: Scalable Vector Graphics.

> [!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` `SVG` `GIF`

### [outputQuality](https://runware.ai/docs/models/openai-gpt-image-2-6-sunburst#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/openai-gpt-image-2-6-sunburst#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/openai-gpt-image-2-6-sunburst#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/openai-gpt-image-2-6-sunburst#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/openai-gpt-image-2-6-sunburst#request-safety)

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

Content safety checking configuration for image generation.

#### [checkContent](https://runware.ai/docs/models/openai-gpt-image-2-6-sunburst#request-safety-checkcontent)

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

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

### [ttl](https://runware.ai/docs/models/openai-gpt-image-2-6-sunburst#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/openai-gpt-image-2-6-sunburst#request-includecost)

- **Type**: `boolean`

Include task cost in the response.

### [numberResults](https://runware.ai/docs/models/openai-gpt-image-2-6-sunburst#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.

### [maskImage](https://runware.ai/docs/models/openai-gpt-image-2-6-sunburst#request-inputs-maskimage)

- **Path**: `inputs.maskImage`
- **Type**: `string`
- **Formats**: `png`
- **File size**: `up to 50 MB`

Image used to specify which areas of the input image should be edited (UUID, URL, Data URI, or Base64). Must have the same width and height as the first reference image. White marks the area to edit, black is preserved. The model treats the mask as guidance rather than a hard boundary, so edits can extend past its shape. With several reference images the mask applies to the first.

### [referenceImages](https://runware.ai/docs/models/openai-gpt-image-2-6-sunburst#request-inputs-referenceimages)

- **Path**: `inputs.referenceImages`
- **Type**: `array of strings`
- **Min items**: `1`
- **Max items**: `16`
- **File size (each)**: `up to 50 MB`

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

### [markupImage](https://runware.ai/docs/models/openai-gpt-image-2-6-sunburst#request-inputs-markupimage)

- **Path**: `inputs.markupImage`
- **Type**: `string`
- **Formats**: `png`
- **Area**: `up to 16.78 MP`
- **File size**: `up to 50 MB`

Image with marks drawn over the first reference image to show where to edit (UUID, URL, Data URI, or Base64). Must have the same width and height as the first reference image and contain transparent pixels. With a markup image, the first reference image can have at most 16,777,216 pixels.

**Core Parameters**

Primary parameters that define the task output.

### [model](https://runware.ai/docs/models/openai-gpt-image-2-6-sunburst#request-model)

- **Type**: `string`
- **Required**: true
- **Value**: `openai:gpt-image@2.6-sunburst`

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/openai-gpt-image-2-6-sunburst#request-positiveprompt)

- **Type**: `string`
- **Required**: true
- **Min**: `2`
- **Max**: `32000`

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/openai-gpt-image-2-6-sunburst#request-width)

- **Type**: `integer`
- **Min**: `16`
- **Max**: `3840`
- **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/openai-gpt-image-2-6-sunburst#request-height)

- **Type**: `integer`
- **Min**: `16`
- **Max**: `3840`
- **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)

**Settings**

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

### [background](https://runware.ai/docs/models/openai-gpt-image-2-6-sunburst#request-settings-background)

- **Path**: `settings.background`
- **Type**: `string`
- **Default**: `auto`

Background handling in generated images.

**Allowed values**:

- `auto` Automatically determines the most appropriate background based on content.
- `opaque` Solid background, no transparency.
- `transparent` Transparent background areas where applicable.

### [gif](https://runware.ai/docs/models/openai-gpt-image-2-6-sunburst#request-settings-gif)

- **Path**: `settings.gif`
- **Type**: `object (2 properties)`

Animated GIF output options.

#### [frameCount](https://runware.ai/docs/models/openai-gpt-image-2-6-sunburst#request-settings-gif-framecount)

- **Path**: `settings.gif.frameCount`
- **Type**: `integer`
- **Min**: `2`
- **Max**: `16`

Number of frames in the animated GIF.

#### [frameDuration](https://runware.ai/docs/models/openai-gpt-image-2-6-sunburst#request-settings-gif-frameduration)

- **Path**: `settings.gif.frameDuration`
- **Type**: `float`
- **Min**: `0.02`
- **Max**: `655.35`
- **Step**: `0.01`

Display time of each GIF frame, in seconds.

### [moderation](https://runware.ai/docs/models/openai-gpt-image-2-6-sunburst#request-settings-moderation)

- **Path**: `settings.moderation`
- **Type**: `string`
- **Default**: `auto`

Content moderation level.

**Allowed values**:

- `auto` Standard filtering that limits creating certain categories of potentially age-inappropriate content.
- `low` Less restrictive filtering.

### [quality](https://runware.ai/docs/models/openai-gpt-image-2-6-sunburst#request-settings-quality)

- **Path**: `settings.quality`
- **Type**: `string`
- **Default**: `auto`

Image quality level.

**Allowed values**: `auto` `max` `xhigh` `high` `medium` `low`

### [svg](https://runware.ai/docs/models/openai-gpt-image-2-6-sunburst#request-settings-svg)

- **Path**: `settings.svg`
- **Type**: `object (2 properties)`

SVG output options.

#### [model](https://runware.ai/docs/models/openai-gpt-image-2-6-sunburst#request-settings-svg-model)

- **Path**: `settings.svg.model`
- **Type**: `string`
- **Default**: `openai:gpt@5.4-mini`

Language model that writes the SVG.

**Allowed values**: `openai:gpt@5.4-mini` `openai:gpt@5.4` `openai:gpt@5.5` `openai:gpt@5.6-terra` `openai:gpt@6-luna` `openai:gpt@6-astra`

#### [thinkingLevel](https://runware.ai/docs/models/openai-gpt-image-2-6-sunburst#request-settings-svg-thinkinglevel)

- **Path**: `settings.svg.thinkingLevel`
- **Type**: `string`

Controls the depth of internal reasoning the model performs before generating a response.

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

### [thinkingLevel](https://runware.ai/docs/models/openai-gpt-image-2-6-sunburst#request-settings-thinkinglevel)

- **Path**: `settings.thinkingLevel`
- **Type**: `string`
- **Default**: `high`

Controls the depth of internal reasoning the model performs before generating a response.

**Allowed values**: `low` `medium` `high`

**Features**

Standalone addons and post-processing features.

### [watermark](https://runware.ai/docs/models/openai-gpt-image-2-6-sunburst#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/openai-gpt-image-2-6-sunburst#request-advancedfeatures-watermark-text)

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

Watermark text.

#### [image](https://runware.ai/docs/models/openai-gpt-image-2-6-sunburst#request-advancedfeatures-watermark-image)

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

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

#### [displayPosition](https://runware.ai/docs/models/openai-gpt-image-2-6-sunburst#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/openai-gpt-image-2-6-sunburst#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/openai-gpt-image-2-6-sunburst#request-advancedfeatures-watermark-fontcolor)

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

Text color in hex format.

#### [bgColor](https://runware.ai/docs/models/openai-gpt-image-2-6-sunburst#request-advancedfeatures-watermark-bgcolor)

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

Background color in hex format.

## Response Parameters

### [taskType](https://runware.ai/docs/models/openai-gpt-image-2-6-sunburst#response-tasktype)

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

Identifier for the type of task this response belongs to.

### [taskUUID](https://runware.ai/docs/models/openai-gpt-image-2-6-sunburst#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/openai-gpt-image-2-6-sunburst#response-imageuuid)

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

UUID of the output image.

### [imageURL](https://runware.ai/docs/models/openai-gpt-image-2-6-sunburst#response-imageurl)

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

URL of the output image.

### [imageBase64Data](https://runware.ai/docs/models/openai-gpt-image-2-6-sunburst#response-imagebase64data)

- **Type**: `string`

Base64-encoded image data.

### [imageDataURI](https://runware.ai/docs/models/openai-gpt-image-2-6-sunburst#response-imagedatauri)

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

Data URI of the output image.

### [seed](https://runware.ai/docs/models/openai-gpt-image-2-6-sunburst#response-seed)

- **Type**: `integer`

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

### [NSFWContent](https://runware.ai/docs/models/openai-gpt-image-2-6-sunburst#response-nsfwcontent)

- **Type**: `boolean`

Flag indicating if NSFW content was detected.

### [cost](https://runware.ai/docs/models/openai-gpt-image-2-6-sunburst#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)

![Subsea Cable Landing Wall Art](https://assets.runware.ai/examples/openai-gpt-image-2-6-sunburst/ee240ebd-30f2-4279-ae0b-2c8e52f647df.jpg)

```json
{
  "taskType": "imageInference",
  "taskUUID": "a52fcc5a-6034-4f9c-a06f-0e02ddb7bdfc",
  "model": "openai:gpt-image@2.6-sunburst",
  "positivePrompt": "Create a finished landscape wall-art illustration showing a dramatic cutaway of a subsea fiber-optic cable landing on a remote volcanic coast. Compose the scene as one continuous panorama divided naturally by the ocean surface: above water, a compact concrete landing station sits among wind-bent grasses on a dark basalt headland, with a narrow maintenance road, small utility structures, distant sea stacks, low clouds, and a pale moon near the horizon; below water, reveal the cable descending through translucent teal waves, crossing layered rock shelves and a dense kelp forest before continuing into the deep ocean. Include precise, believable engineering details: the armored shore cable entering a protective conduit, articulated cable joints, anchor points, a cylindrical optical repeater farther offshore, and a compact yellow inspection ROV hovering nearby with two soft beams illuminating suspended particles. Add marine life at varied scales—silver schooling fish threading through kelp, an octopus tucked into basalt, anemones and cold-water corals on the rocks, and two distant whales rendered as quiet silhouettes in the open water.\n\nArt direction: sophisticated contemporary printmaking that combines reduction linocut shapes with fine copperplate-style engraved detail. Use crisp layered contours, controlled crosshatching, subtle paper grain, and carefully registered blocks of color. Palette limited to midnight indigo, deep petrol blue, oxidized copper, muted coral, sea-glass green, warm ivory, and small accents of signal yellow on the ROV. The image should feel contemplative, intelligent, and premium rather than like a textbook diagram. Strong visual flow follows the cable in a sweeping diagonal from the landing station at upper left toward the deep sea at lower right. Balance intricate detail with calm areas of open water and sky. Keep all essential subjects comfortably inside the edges for print cropping. Flat, borderless finished artwork only—no frame, room mockup, labels, captions, logos, symbols, map keys, or other lettering.",
  "width": 1536,
  "height": 1024
}
```

---

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

![Adaptive Rainwear Fashion Lookbook](https://assets.runware.ai/examples/openai-gpt-image-2-6-sunburst/1d5633a8-ae41-4d13-8019-e8f24e7a1842.jpg)

```json
{
  "taskType": "imageInference",
  "taskUUID": "f067590b-7e07-4bcf-8a0b-002338d3d9bc",
  "model": "openai:gpt-image@2.6-sunburst",
  "positivePrompt": "Create a finished 1024x1536 portrait fashion lookbook photograph for an adaptive outerwear collection. Combine the two references into a new, coherent image rather than reproducing either source. Use reference image 1 specifically for the adult model's recognizable facial identity, hair, body proportions, natural seated posture, and the proportions and structure of her manual wheelchair; do not copy her reference clothing or studio background. Use reference image 2 specifically for the drained municipal pool architecture, mint-and-cream palette, crisp winter daylight, direct-flash character, and restrained film-grain art direction; do not copy its exact camera position.\n\nShow the model seated confidently in her wheelchair on the floor of the empty drained pool, photographed full length in a refined three-quarter view. Keep her entire outfit, hands, feet, wheels, push rims, and chair frame clearly visible and anatomically coherent. She wears a premium adaptive rainwear look: a sculptural rust-orange waterproof trench cut specifically for a seated silhouette, cropped slightly higher at the front and longer at the sides, with a clean magnetic storm placket, asymmetric cape shoulder, articulated elbows, discreet side-entry pockets positioned within seated reach, and a removable cobalt-blue hood folded neatly at the collar. Pair it with cobalt waterproof trousers featuring shaped knees, smooth seated drape, subtle ankle zips, and matte charcoal weatherproof boots. Include realistic technical fabric texture, taped-seam construction, precise tailoring, small tonal hardware, and believable folds created by the seated pose. No visible brand names or logos.\n\nHer pose should feel composed and self-possessed rather than athletic: torso gently angled toward camera, one hand resting on a push rim and the other lightly touching the trench lapel, direct calm eye contact. Position her slightly below center with generous architectural negative space above, while keeping the clothing large enough to inspect as a catalog garment. A stainless-steel pool ladder and tiled depth markers without readable lettering may appear as restrained background geometry. Use the mint tiles to contrast with the rust trench and cobalt trousers.\n\nArt direction: high-end independent fashion magazine meets practical apparel catalog, cool architectural daylight with a controlled hard flash that creates a clean shadow and reveals fabric construction, crisp natural skin texture, accurate garment edges, subtle analog grain, restrained color grading, realistic optics, vertical composition, no artificial blur. The image must read as one intentional on-location fashion photograph, not a collage or composite. No text, captions, borders, logos, watermarks, extra people, duplicate limbs, extra wheels, malformed hands, or distorted wheelchair geometry.",
  "width": 1024,
  "height": 1536,
  "inputs": {
    "referenceImages": [
      "https://assets.runware.ai/assets/inputs/3f2337d4-b55f-4a89-a463-c52f508c9e00.jpg",
      "https://assets.runware.ai/assets/inputs/553f03db-2fcc-4a53-9a4e-d76a848e8a67.jpg"
    ]
  }
}
```

---

### inpaint (inpaint)

![Cave Rescue Equipment Sticker Pack](https://assets.runware.ai/examples/openai-gpt-image-2-6-sunburst/41fefb8a-52e3-49d2-b59f-7ac1bea4afa0.jpg)

```json
{
  "taskType": "imageInference",
  "taskUUID": "11d19781-e1bc-4800-b9b2-c374fc42cb2d",
  "model": "openai:gpt-image@2.6-sunburst",
  "positivePrompt": "Edit the masked subject only. Replace the entire blank white sheet with a professionally printed, kiss-cut vinyl sticker pack designed for volunteer cave-rescue teams to label rugged equipment cases. Preserve the sheet’s exact outer silhouette, size, position, rounded corners, thickness, camera angle, perspective, contact shadow, highlights, and relationship to the tabletop. Preserve the surrounding charcoal-gray tabletop, lighting, framing, and composition exactly.\n\nCreate a cohesive collection of 12 distinct die-cut stickers arranged efficiently across the sheet with realistic gaps and visible kiss-cut outlines. Use bold, highly legible rescue pictograms: coiled static rope with carabiner; climbing helmet with headlamp; compact field radio; trauma medical pouch with a simple cross; waterproof survey map with cave passage line; battery pack with charge bars; water bottle; air-quality detector; rescue litter; mechanical ascender; gloved hand giving an OK signal; and a circular team emblem combining a cave entrance, rope, and headlamp beam. Include short functional labels only where useful: “ROPE”, “COMMS”, “MEDICAL”, “LIGHT”, “AIR”, “MAP”, and “CAVE RESCUE”. Spell every label exactly.\n\nArt direction: crisp utilitarian pictograms influenced by 1970s European mountain-safety graphics and modern wayfinding systems, with sturdy geometric shapes, simplified linework, slightly rounded corners, and excellent recognition at small sizes. Restrained high-visibility palette of fluorescent safety orange, deep midnight navy, warm off-white, and small reflective-silver accents. Give each sticker a thick off-white vinyl border and subtle satin lamination highlights consistent with the original studio light. Vary the sticker silhouettes—circles, rounded rectangles, shield shapes, and contour-cut equipment forms—while maintaining one unified visual system. The result should look like a real premium printed sticker sheet ready for manufacture, not loose stickers, not a flat digital layout, and not a presentation mockup. Do not alter any pixels outside the masked sheet.",
  "width": 1280,
  "height": 1280,
  "inputs": {
    "referenceImages": [
      "https://assets.runware.ai/assets/inputs/b1b71fdf-b551-47ae-b375-01f1a990ad99.jpg"
    ],
    "maskImage": "https://assets.runware.ai/assets/inputs/b4a6bb34-8df1-48ed-9574-c67a565bbf57.png"
  }
}
```

---

### text-to-vector (text-to-vector)

![Nocturnal Moth Survey Festival Poster](https://assets.runware.ai/examples/openai-gpt-image-2-6-sunburst/4ff0ac25-6b83-4535-bbd7-5b31a41efeff.svg)

```json
{
  "taskType": "imageInference",
  "taskUUID": "ec490828-0b9e-4dd7-b202-74ed38ece9a8",
  "outputFormat": "SVG",
  "model": "openai:gpt-image@2.6-sunburst",
  "positivePrompt": "Create a finished vertical event poster for an urban ecology festival and citizen-science moth survey. Flat vector-suitable artwork with precise geometry, crisp edges, clean negative space, solid fills, and screen-print character. No gradients, shadows, photographic texture, or mockup presentation.\n\nCentral illustration: a large, symmetrical luna moth viewed from above, wings fully spread. Construct the moth from bold, simplified shapes with a few elegant internal wing lines. Place a small glowing survey lantern at the center of its body. Integrate subtle urban details into the wings using minimal geometric silhouettes: apartment windows, native leaves, stars, and tiny moth flight paths. The composition should feel scientific, nocturnal, welcoming, and sophisticated rather than spooky.\n\nUse a tightly controlled four-color palette: deep midnight navy background, warm cream type and linework, electric citron-green moth wings, and restrained coral-orange accents. Preserve strong contrast and generous margins. Make all elements suitable for clean vector tracing and spot-color printing.\n\nTypography is fundamental to the poster. Use bold condensed sans-serif display lettering with clean modern supporting type. Render the following wording exactly, clearly, and in this hierarchy:\n\nMIDNIGHT INDEX\nA NIGHT MOTH BIOBLITZ\nSATURDAY 18 JULY · 8 PM–1 AM\nRIVERGLASS BOTANIC YARD\nLIGHT TRAPS · FIELD TALKS · SPECIES COUNT\nFREE · ALL AGES · BRING A TORCH\n\nPlace “MIDNIGHT INDEX” large across the top, the subtitle directly beneath it, and the date and venue in a compact information block near the bottom. Set the activities line as a neat divider or badge row, with the final admission line clearly visible at the base. Include no other words, logos, QR codes, borders, sponsor marks, or decorative pseudo-text. Deliver only the standalone poster artwork, straight-on and filling the canvas.",
  "width": 1024,
  "height": 1536
}
```

---

### transparent-background (transparent-background)

![Oyster Farm Children’s Comic Panel](https://assets.runware.ai/examples/openai-gpt-image-2-6-sunburst/6aebdc4e-8521-48f1-9276-e8c87aa2e49e.png)

```json
{
  "taskType": "imageInference",
  "taskUUID": "33432745-6339-4135-9318-8237e6dfc835",
  "outputFormat": "PNG",
  "model": "openai:gpt-image@2.6-sunburst",
  "positivePrompt": "Create a finished square, wordless children’s comic spot panel about sustainable oyster farming. Show one clearly defined isolated subject: a Black woman oyster farmer in her 30s, full body in a lively three-quarter pose, carefully fastening an identification tag to the corner of one rectangular mesh oyster-nursery basket held securely against her hip. Her expression is focused, capable, and warmly encouraging. She wears accurate working gear: mustard-yellow waterproof bib overalls over a deep navy shirt, seafoam rubber deck boots, gray work gloves, and a coral knit cap. Keep the equipment practical and believable for a modern coastal shellfish nursery.\n\nArt direction: sophisticated mid-century children’s editorial illustration with crisp ink contours, simplified geometric shapes, expressive hands and face, and restrained risograph-style color registration. Limited palette of deep navy, seafoam green, coral red, mustard yellow, and warm cream. Add subtle halftone and dry-brush texture only inside the character and equipment shapes. Strong readable silhouette, clean closed edges, clear separation between limbs and basket, polished print-ready finish suitable for readers ages 8–11.\n\nThe character and held basket must form one compact, self-contained cutout centered in the square with comfortable transparent margin on every side. True transparent background everywhere outside the subject; no environmental backdrop, water, dock, sky, ground plane, border, panel frame, cast shadow, decorative objects, lettering, captions, speech balloons, logos, or watermark.",
  "width": 1024,
  "height": 1024,
  "settings": {
    "background": "transparent"
  }
}
```

---

### text-to-gif (text-to-gif)

![Geothermal Plant Director Looping Portrait](https://assets.runware.ai/examples/openai-gpt-image-2-6-sunburst/193dd442-464c-46ef-b3e6-c5c6c6388fed.gif)

```json
{
  "taskType": "imageInference",
  "taskUUID": "ff1c3b1e-ab3f-4b61-8502-6a7ea07f675b",
  "outputFormat": "GIF",
  "model": "openai:gpt-image@2.6-sunburst",
  "positivePrompt": "Create a short, seamlessly looping animated professional portrait in a refined editorial photography style. Subject: a Māori woman in her early 50s who is the operations director of a geothermal power station, photographed from mid-chest upward. She has warm brown skin, silver-streaked dark hair cut in a neat chin-length bob, subtle natural makeup, and a calm, capable expression with direct eye contact. She wears a beautifully tailored charcoal work jacket over a muted mineral-green blouse, small pounamu stud earrings, and no visible branding. Her posture is upright and relaxed.\n\nSet her inside a modern geothermal turbine hall, rendered as a softly defocused industrial background with pale concrete, brushed steel pipework, and restrained amber equipment lights. Use soft directional window light from camera left, gentle facial modeling, realistic catchlights, natural skin texture, and a faint warm rim light separating her from the cool background. Color palette: charcoal, pumice gray, mineral green, weathered copper, and quiet amber. Premium corporate editorial art direction, credible workplace context, dignified rather than glamorous.\n\nCompose vertically as a clean professional headshot with the subject centered slightly right, eyes near the upper third, both shoulders visible, and generous uncluttered space around the head. Keep the camera completely locked. Motion must be limited and clearly readable: one natural blink, an almost imperceptible breath that raises and lowers the shoulders, and a subtle pulse in two distant amber indicator lights. Hair, clothing, facial features, framing, focus, and background geometry remain stable. Begin and end on the exact same calm expression, posture, lighting, and indicator-light state so the final frame transitions invisibly back to the first. No camera movement, cuts, morphing, speaking, exaggerated expression, text, logos, badges, watermarks, or interface elements.",
  "settings": {
    "gif": {
      "frameCount": 8,
      "frameDuration": 0.12
    }
  }
}
```