Updated

GPT Image 2.5

OpenAI introduced an image model with sharper details, more precise editing, and faster generation.

By No Code MBA · Facts checked

Provider
OpenAI
Announcement date
Status at last check
Generally available
Model family
GPT Image
Preceding model
GPT Image 2
Capabilities
Image, Text

What changed

GPT Image 2.5 powers ChatGPT Images 2.5 and ships in the API as gpt-image-2.5-flare, which OpenAI calls its fastest model for everyday high-quality images. OpenAI cites faster generation, more consistent details across edits, and edits that change only the requested area. The API adds xhigh and max quality levels, new sizes and an opaque background option.

Access and availability

Use gpt-image-2.5-flare (snapshot gpt-image-2.5-flare-2026-09-08) on the Images generations, edits and Batch endpoints; Responses and Chat Completions are not supported. In ChatGPT it is rolling out to all users, including ChatGPT Work and Codex.

Official access documentation

Pricing

GPT Image 2.5 Flare: $5 per million text input tokens ($1.25 cached), $8 per million image input tokens ($2 cached), and $30 per million image output tokens ($15 with Batch)

Official pricing

What you can build

Use repeated edits to evaluate whether a product image keeps its key details intact.

Limitations

OpenAI did not publish speed figures. Default rate limits start at 20 images per minute on the Build tier, and ChatGPT usage caps by plan were not stated.

Sources and release history