OpenAI Debuts ChatGPT Images 2.5 with Sketching and 50% Faster Generation
The new model introduces direct drawing tools and reduced latency to shift AI art toward a professional creative workflow.
OpenAI has released ChatGPT Images 2.5, a state-of-the-art image model designed to provide sharper details and more precise editing capabilities. The update marks a significant step in the company's effort to move beyond simple prompt-based generation toward a more controlled, collaborative creative process.
The new model delivers a substantial performance boost, with image generation latency reduced by up to 50% compared to the previous Images 2.0 version. A centerpiece of the update is the 'Sketch' feature, which allows users to draw directly within the ChatGPT interface to provide a visual guide for the final output. To support developers, OpenAI also introduced two API-specific models: GPT-Image-2.5 Flare, which balances quality and speed, and GPT-Image-2.5 Sunburst, designed for high-precision premium workflows. The update is currently available to all ChatGPT, ChatGPT Work, and Codex users across web, mobile, and desktop platforms.
Iterating on Fidelity
This release arrives approximately five months after the launch of ChatGPT Images 2.0. The rapid iteration reflects OpenAI's focus on improving fidelity, particularly when users work with reference photos. By introducing templates for popular formats and enhancing multi-turn consistency, the company is addressing a primary pain point for AI artists: the difficulty of maintaining a consistent subject or style across multiple edits. The goal is to provide more granular control over the editing process, reducing the reliance on the randomized nature of traditional AI generation.
Shifting to Creative Workflows
The introduction of sketching and high-fidelity reference preservation signals a strategic shift in how AI image tools are positioned. Rather than serving as a novelty for generating artistic imagery, Images 2.5 targets professional creators and developers who need precise control over brand assets and UI concepts. By integrating direct visual input, OpenAI is transforming the tool into a functional part of a professional design pipeline, where the AI acts as a collaborator that follows a specific blueprint rather than an autonomous generator.
The Road Ahead
As OpenAI continues to refine its image capabilities, the industry will be watching how these professional-grade tools impact the broader design market. While the current update focuses on speed and precision, it remains to be seen how these tools will integrate with other multimodal capabilities in future updates. For now, the focus remains on bridging the gap between a text prompt and a finished, professional-grade visual asset.