OpenAI has announced ChatGPT Images 2.5, a state-of-the-art image generation model that introduces sharper details, more precise editing, and faster creative workflows. The company notes that people already create more than 3 billion images across ChatGPT Images and GPT-Image models in the API every week. This update is designed to help both everyday users and developers move from rough concepts to polished visuals with greater control.
Improved Image Fidelity and Faster Generation
Images 2.5 produces more natural lighting and richer textures while better preserving subjects from reference photos. Generation latency has been reduced by up to 50 percent compared with Images 2.0, enabling faster refinement of ideas. The model also follows editing instructions more reliably across multiple turns, so earlier changes remain consistent as new requests build on prior work.
Precision Editing and Multi-Turn Consistency
The model makes focused edits more accurate by changing only what a user has asked for and keeping surrounding details intact. This is especially important for complex subjects and backgrounds where small adjustments should not disturb the rest of the image. In longer conversations, quality remains stable across successive edits, which supports production workflows that require targeted changes without rebuilding assets.
New Creative Controls in ChatGPT
OpenAI is introducing Sketch, a feature that lets users draw directly inside ChatGPT to use as a visual guide for the final image. Templates are also available for popular formats such as posters, flyers, and product photos, reducing the need to start from a blank canvas. These additions aim to give users more control over composition and style while lowering the barrier for creative experimentation.
Commenting and Prompt Sharing
Users can now place comments directly on images to request more focused edits, making feedback easier to communicate. When sharing an image, they may also include the original prompt so other people can try the same idea with their own photos and details. This prompt sharing feature is intended to help creative ideas spread while allowing each person to personalize the result.
Developer Models for API Workflows
For developers, OpenAI is releasing two new models in the Images API. GPT-Image-2.5 Flare brings the same improvements in quality, editing, and speed and serves as the default choice for most applications, delivering higher-quality images than GPT-Image-2 at 50 percent lower latency. GPT-Image-2.5 Sunburst offers an extra level of precision for detailed creative work with longer generation times, suitable for premium visual workflows.
Intelligence, Style and Safety
Images 2.5 is better at understanding complex visual instructions and translating them into coherent results, including more accurate real-world information and complex layouts with transparent backgrounds. It also reflects artistic styles more faithfully, making it useful for on-brand creative assets, UI concepts, and presentation visuals. Safety safeguards include prompt and image checks, C2PA metadata, and invisible watermarking to help identify content made with OpenAI tools.
Availability and Pricing
Images 2.5 is rolling out today to ChatGPT, ChatGPT Work, and Codex users across all tiers on desktop, mobile, and web. The two new API models, GPT-Image-2.5 Flare and GPT-Image-2.5 Sunburst, are available for developers, with pricing details published by OpenAI. This broad rollout supports both everyday creative projects and high-volume production environments.
OpenAI's latest update combines faster generation, stronger reference fidelity, and more intuitive editing tools in a single release. By offering dedicated API models and consumer-facing features such as Sketch, templates, and prompt sharing, the company is addressing both individual creators and business teams. ChatGPT Images 2.5 represents a meaningful advance in making AI-generated visuals more accurate, accessible, and production-ready.