O

OpenAI: gpt-image-2.6-sunburst

gpt-image-2.6-sunburst
Vision4 img in
Released Nov 5, 2026Updated Nov 5, 2026

GPT Image 2.6 Sunburst is OpenAI's high-quality, precision-oriented tier of the GPT Image 2.6 series, released in 2026 and positioned alongside Flare. It accepts image input and produces image output through the Images API, and it is also available in ChatGPT and Codex across all subscription levels on desktop, mobile, and web. The model supports an image markup feature that lets users specify exact areas to change, and it maintains character, product, and brand details across edits. The model exposes a complexity parameter that adjusts how much planning it does to satisfy a request, plus quality settings selectable from low to high. Higher-quality generation is produced at the medium or high quality settings, and SVG export makes designs directly usable on websites, apps, and print. Generation speed is improved by 50% over the previous model, and image consistency between generation and editing is stronger. Sunburst suits detailed work where editing accuracy matters more than generation speed: animating brand mascots and characters, building GIFs that turn product shots into engaging loops, precision editing while preserving character, product, and brand details, professional visual production such as posters and product imagery, and creative workflows with multiple editing rounds. It is also available via the API as GPT-Image-2.6 Sunburst and has been added to Figma Weave.

Inputs used for the providers' own purposes

At least one provider of this model uses your inputs for its own purposes, for example to train AI models. Do not enter personal data of third parties or sensitive data. You can object to this use at any time in your account settings or via "Manage consents" in the chat.

Read the privacy policy

Pricing

Price per request
$0.003/ image
Compatible endpoints image-generation, openaiVendor OpenAIProvider Routerah

Uptime

Performance

Loading performance data...

Usage & Ranking

Loading usage...

Supported parameters

All providers = every upstream serving this model supports it. Some providers = depends on which upstream handles the request. Default = the value sent when you leave the parameter unset.

ParameterProvidersDefault
aspect_ratioSome providers-
backgroundSome providers-
input_referencesSome providers-
nSome providers-
output_compressionSome providers-
output_formatSome providers-
qualitySome providers-

Frequently asked questions

Similar models