Overview
- The new model, released Tuesday across ChatGPT, Codex and the API, now serves as the default for image generation on the platform.
- OpenAI says it follows complex instructions more closely, places objects with precision, and renders on‑image text cleanly in languages including Japanese, Korean, Chinese and Hindi.
- Reporting on the launch notes support for aspect ratios from 3:1 to 1:3 and up to eight images per prompt, which speeds tasks like storyboards and multi‑format campaigns.
- Advanced “Thinking” features that can search the web and double‑check outputs are available to Plus, Pro, Business and Enterprise subscribers.
- OpenAI keeps content safeguards and monitoring in place, and the release arrives as competitors such as Canva and Google’s Nano Banana 2 push pro‑grade image tools.