ChatGPT Images 2.0

ChatGPT Images 2.0 (model ID gpt-image-2) is OpenAI's next-generation image model, launched April 21, 2026 as the default image generator in ChatGPT and the OpenAI API. It replaces DALL·E 2 and DALL·E 3, both of which retire on May 12, 2026.

The defining capability is native reasoning before generation: the model "thinks" through the prompt — composition, layout, character coherence, and constraints — before rendering, leading to noticeable gains on the kinds of details image models historically struggle with. Small text, iconography, UI elements, dense compositions, and subtle stylistic constraints are all rendered with markedly higher fidelity. Output supports resolutions up to 2K and aspect ratios from 3:1 to 1:3.

Two modes are available. Instant Mode brings core quality improvements to every ChatGPT user, including the free tier. Thinking Mode — which adds web search, layout reasoning, multi-image batching, and output verification — is restricted to Plus ($20/mo), Pro ($200/mo), Business and Enterprise tiers. Thinking Mode produces up to 8 coherent images per prompt with character and object continuity across the batch, making it especially useful for storyboards, social-media campaigns, and product imagery that needs to look like a series.

Within 12 hours of launch, ChatGPT Images 2.0 took the #1 spot across every category on the Image Arena leaderboard by a +242-point margin. For ChatGPT users who want the best image quality without leaving the chat — and for developers who want a single API for both text and image generation — gpt-image-2 is the new default.