OpenAI image model

GPT Image 2.6. Sunburst for detail, Flare for speed.

OpenAI's newest image model in two tiers: Sunburst for dense, detailed frames and Flare for fast drafts. Start from a sentence, or attach a photo and describe the change.

Try GPT Image 2.6->See all models

Text and image input
Up to 10 reference images
8 aspect ratios plus auto
Quality from low to max
Transparent background

GPT Image 2.6 is OpenAI's model for making and editing images, released as two tiers. Sunburst is tuned for detail: busy compositions, text-heavy layouts, and edits that keep the same details from one pass to the next. Flare is tuned for speed, for quick iteration from text or from an image. Both take a text prompt and reference images, with the same eight aspect ratios plus auto.

On Treza both sit in the Image Generation node's model picker, with a Quality setting from low to max and a transparent background option. Flare also runs in chat. Draft on Flare at low quality, then re-run the keeper on Sunburst at high or max. A finished still wires into the First frame port of a video node, and the whole pipeline publishes as one API.

Text that reads as written

Menus, signs, and labels come back spelled the way the prompt spells them, so a storefront or a package mockup does not need a second pass.

Edits that keep the details

Attach an image and describe the change. The subject, its texture, and the framing carry over, so a product stays the same product across a set.

Transparent backgrounds

Both tiers can return a transparent PNG, a product cutout or an icon ready to place over any layout.

Six quality levels

Low, medium, high, extra high, and max on both tiers. Each image is charged for the quality it rendered at, so drafts stay cheap and only the final costs more.

What to make with GPT Image 2.6

Prompts to try, and what to expect back. Each runs on an Image Generation node with GPT Image 2.6 Sunburst or GPT Image 2.6 Flare picked in the model menu.

“A cafe chalkboard menu listing Espresso, Americano, Latte, and Flat White, on Sunburst”

A photographic still with every item spelled as written.

“This shirt (attached) folded on a white marble counter in a bright studio”

The same shirt, restaged, with its fabric and stitching intact.

“Four quick concepts for a cooking video thumbnail, Flare at low quality”

A batch of drafts to choose a direction from.

“The chosen concept again on Sunburst at max quality, 16:9”

The keeper, finished with more detail.

“This mug (attached) as a cutout on a transparent background”

A PNG ready to place on any layout.

“A poster for a jazz night with the date, the venue, and three band names”

A dense layout with the copy in place.

GPT Image 2.6 is one node on a canvas

Chain it with language, image, and video models, then publish the whole pipeline as an API. See how a full AI video pipeline fits together.

Step 01
Draft the brief
A language model node expands a one-line idea into a detailed, on-brand prompt before it ever reaches the model.
Step 02
Generate with GPT Image 2.6
Run variants in parallel, compare against other models on the same prompt, and keep every result in run history.
Step 03
Publish it as an API
The pipeline becomes a versioned endpoint. Your product calls it with the typed /invoke API or any OpenAI SDK.

GPT Image 2.6 as an API, priced per image

The GPT Image 2.6 API

Build a pipeline with GPT Image 2.6 on a node and publish it. It becomes a versioned HTTP endpoint: the typed /invoke route enqueues a run and returns a run id to poll, and an OpenAI-compatible route works with SDKs you already use. Agents connect over MCP the same way. Your prompt engineering ships inside the pipeline, so every caller gets your quality, and swapping the model later does not change the API your product calls.

GPT Image 2.6 pricing

Prepaid credits, no subscription. Each image is charged at GPT Image 2.6's own rate from your balance, shown before you run, and a failed generation costs nothing. Credit packs start at $5 and never expire, which suits spiky production schedules. Run history records the cost of every generation, per node, so spend is auditable rather than a surprise.

GPT Image 2.6, answered

What is the difference between GPT Image 2.6 Sunburst and Flare?

Sunburst is the detail tier, for dense compositions and text-heavy layouts. Flare is the speed tier: in our tests it finished a low quality image in about 6 seconds and a max quality one in about 40, where Sunburst took about 17 and 166. Both have the same settings and are charged the same per image at each quality level.

How is GPT Image 2.6 different from GPT Image 2.5?

Both come as Sunburst and Flare with the same quality ladder, transparent backgrounds, and aspect ratios. In our tests 2.6 rendered its low and medium levels with more image tokens than 2.5, so those levels cost more per image, while high, extra high, and max cost the same.

Can I use GPT Image 2.6 in chat?

Flare, yes: ask for it by name and chat renders with it. Sunburst runs on the canvas, where its longer renders at the top quality levels do not hold up a conversation.

How much does GPT Image 2.6 cost on Treza?

Treza uses prepaid credits with no subscription. Each image is charged at GPT Image 2.6's rate from your balance, and only successful generations are charged. Credit packs start at $5 and never expire.

Can I call GPT Image 2.6 through an API?

Yes. Build a pipeline that uses GPT Image 2.6 and publish it as a versioned HTTP endpoint. Call the typed /invoke API from your product, or point any OpenAI SDK at the OpenAI-compatible endpoint. If a better model ships later, swap it on the node without changing the API your product calls.