About GPT Image 2.6 Sunburst
GPT Image 2.6 Sunburst is OpenAI's image generation and editing model for detailed work, keeping details consistent across edits at medium and high quality, and about 50% faster than GPT Image 2.5 Sunburst.
It generates from a prompt or edits up to 16 reference images, with markup, mask inpainting, transparent backgrounds, sizes up to 4K (3840x2160) and five quality tiers from low to max. From a prompt alone it can also render an animated GIF of 2 to 16 frames or a scalable SVG. It suits campaign creative and polished product imagery; GPT Image 2.6 Flare is the faster choice for quick iteration.
See what this model can do
A handful of real prompts and their outputs.
Ready to build with GPT Image 2.6 Sunburst?
Try GPT Image 2.6 Sunburst in the Workbench to prompt it, compare outputs, and iterate on prompts without writing any code. When you're ready to ship, call the same model from our API and build your own apps on top of it.
curl -sSf -X POST https://hub.oxen.ai/api/ai/images/edit \
-H "Content-Type: application/json" \
-H "Authorization: Bearer $OXEN_API_KEY" \
-d '{
"model": "gpt-image-2-6-sunburst",
"lock_unmasked": true,
"aspect_ratio": "16:9",
"resolution": "2K",
"quality": "high",
"output_format": "png",
"frame_count": 9,
"frame_duration_ms": 200,
"svg_model": "gpt-6-sol",
"complexity": "high",
"background": "auto",
"moderation": "low"
}'Request parameters
Every field GPT Image 2.6 Sunburst accepts in the request body. See the API reference for response formats.
https://hub.oxen.ai/api/ai/images/edit| Field | Type | Required | Default | Description |
|---|---|---|---|---|
input_image | array<string>nullable | No | — | Up to 16 images to edit or draw from. |
markup | stringuri · nullable | No | — | Draw on Image 1 to show what to change. Text and arrows are read as instructions. |
mask | stringuri · nullable | No | — | Paint the part of Image 1 to change. |
lock_unmasked | boolean | No | true | Only modify the pixels within the masked region. |
prompt | string | No | — | What to generate, or how to edit the reference images (@Image1, @Image2, ...). Optional with a reference image. |
aspect_ratio | string | No | "16:9" | auto1:116:99:164:33:43:22:321:99:21 |
resolution | string | No | "2K" | 1K2K4K |
quality | string | No | "high" | Higher tiers add detail, take longer and cost more.lowmediumhighxhighmax |
output_format | string | No | "png" | GIF and SVG generate from the prompt alone.pngjpegwebpgifsvg |
frame_count | integer | No | 9 | More frames move more smoothly, each at a smaller size. Range: 2 – 16 |
frame_duration_ms | integer | No | 200 | How long each frame shows. Range: 20 – 2000 |
svg_model | string | No | "gpt-6-sol" | Draws the SVG. Sol adds detail; Luna is faster and cheaper.gpt-6-solgpt-6-luna |
complexity | string | No | "high" | How much the model plans before rendering. Lower is faster, same price.lowmediumhigh |
background | string | No | "auto" | Transparent needs PNG or WebP output.autoopaquetransparent |
moderation | string | No | "low" | Low filters less than auto. OpenAI's usage policies still apply.lowauto |