# Introducing ChatGPT Images 2.5

The new model cuts image generation latency by up to 50% while improving multi-turn editing consistency, subject preservation, and text layout handling.

- Available across desktop, mobile, and web for all ChatGPT, ChatGPT Work, and Codex users
- API developers get two models: default GPT-Image-2.5 Flare and higher-precision GPT-Image-2.5 Sunburst
- Introduces Sketch, allowing users to draw directly in ChatGPT using '@Sketch' to steer generated output
- Adds image templates, direct image commenting for localized edits, and prompt-sharing features
- Improves support for complex layouts, transparent backgrounds, and reference-photo fidelity

## Why it matters

ChatGPT and API users can perform iterative, multi-turn image edits without losing original subjects or degrading image quality over consecutive prompts.

## Sources

- [OpenAI: Introducing ChatGPT Images 2.5](https://openai.com/index/introducing-chatgpt-images-2-5)

---

Summarized by dstilled on 2026-09-08. https://dstilled.ai/story/72527a3f-df9b-465c-a1a3-a40a27b284b8