SoonMaker

ChatGPT Images 2.5 (GPT Image 2.5) — OpenAI's Newest Image Model

ChatGPT Images 2.5 (the GPT Image 2.5 model) is OpenAI's newest image model, released in September 2026. It keeps what made GPT Image 2 useful — strong prompt following and clean in-image text — and adds better preservation of subjects from reference photos, more precise single-element edits, and noticeably faster generation (OpenAI reports up to 50% lower latency). On SoonMaker it runs as the fast Flare tier, so everyday images come back in seconds. It is our default image model: start here for portraits, posters, product shots and edits from a reference photo.

Image model · OpenAI

ChatGPT Images 2.5 (GPT Image 2.5) — OpenAI's Newest Image Model

Sharper detail, faithful edits from your own photos, and faster than GPT Image 2.

Key features

6 highlights

OpenAI's newest image model (September 2026)

Keeps the subject of your reference photo recognisable

Precise single-element edits

Clean, readable text inside images

Faster than GPT Image 2

Our default image model

Specifications

Capabilities &
supported configs

  • ResolutionUp to 1K
  • Aspect ratios1:1, 3:4, 4:3, 16:9, 9:16, 2:3, 3:2

FAQ

Frequently asked

Ready to try ChatGPT Images 2.5?

Start generating with free credits. No credit card required.

Try ChatGPT Images 2.5 Free