ChatGPT Images 2.0 is OpenAI's next-generation AI image generation platform powered by the gpt-image-2 model, launched on April 21, 2026. It represents a major leap forward from the original Images release, introducing thinking capabilities that allow the model to reason before drawing — enabling web search integration, layout reasoning, multi-image batching, and output self-verification. The core breakthrough is dramatically improved text rendering: Images 2.0 can accurately generate readable text in images, handle non-Latin scripts including Japanese, Korean, Hindi, and Bengali, and produce production-level infographics, UI mockups, marketing assets, and multi-panel comic strips. The model supports 2K resolution (up to 2560x1440 experimental), aspect ratios from 3:1 to 1:3, and can generate up to 8 coherent images from a single prompt while maintaining consistent characters and objects across the batch. Two access modes are available: Instant Mode for all ChatGPT users delivers core quality improvements and single-image generation, while Thinking Mode — exclusive to Plus, Pro, Business, and Enterprise subscribers — unlocks web research integration, multi-image batching, layout reasoning, and self-verification. The gpt-image-2 API is also available to developers with usage-based pricing starting at $0.006 per image at low quality and up to $0.211 per high-quality 1024x1024 image. ChatGPT Images 2.0 is integrated across ChatGPT, Codex, and the OpenAI API, positioning it as a replacement for dedicated image editing applications for many professional use cases.