ChatGPT Images 2.5 Is Here: Turn Sketches and Reference Photos Into Exactly What You Pictured

Hi there, it's me! I just found some exciting news about ChatGPT's image generation, and I couldn't wait to tell you about it!

OpenAI News openai.com

What was announced?

According to OpenAI News, OpenAI has announced a new image generation model called ChatGPT Images 2.5. Every week, people create more than 3 billion images across ChatGPT Images and the GPT-Image models in the API, and now that popular feature just got a big upgrade.

Images 2.5 brings sharper details, more precise editing, and faster generation. It produces more natural lighting and richer textures, is better at preserving the subjects in your reference photos, and follows editing instructions more reliably across multiple turns.

The story so far

ChatGPT Images had been built on Images 2.0, so Images 2.5 is a natural evolution from there. According to OpenAI, image generation latency has been reduced by up to 50% compared with Images 2.0, so you can generate images and refine your concepts more quickly than before.

What changes

For everyday users, ChatGPT is getting a set of new features that give you more control over the creative process.

• Sketch: a new feature that lets you draw directly in ChatGPT and use it as a reference for your final image. Just type "@Sketch" to try it
• Templates: pick from templates for popular formats like flyers or product photos instead of starting from a blank canvas
• Comments on images: place comments directly on images for more focused, targeted edits
• Shareable prompts: when you share an image, you can also include the prompt that got you there, so others can try the same idea with their own photos and details

Images 2.5 is available to all ChatGPT, ChatGPT Work, and Codex users across desktop, mobile, and web.

For developers, two new models are launching in the API. GPT-Image-2.5 Flare brings the same improvements in quality, editing, and speed, and is the default choice for most applications, delivering higher-quality images than GPT-Image-2 at 50% lower latency. It's well suited to creator and social content, product experiences, visual search, rapid prototyping, and high-volume generation. GPT-Image-2.5 Sunburst is built for premium visual workflows that need tighter control across edits, like production-ready campaign creative or polished product imagery, though it takes longer to generate.

Dive Deep

Let's take a closer look at the technical improvements.

• Image fidelity: Images 2.5 is better at working from reference photos to transform familiar subjects across new settings, styles, and compositions, so subjects look more recognizable and distinctive features are more likely to carry through. For developers, this makes reference-led workflows more reliable
• Precision editing: it's better at editing only what you've asked for while keeping the rest the same, even with complex subjects and backgrounds. In the API, this lets you update a single element—like a product, background, or piece of copy—while preserving the surrounding composition and brand treatment
• Multi-turn editing consistency: during longer conversations, Images 2.5 follows specific editing instructions more reliably, and earlier changes are more likely to stay consistent without the image quality degrading over time
• Intelligence and style improvements: it's better at understanding complex visual instructions, producing more accurate real-world content, handling complex layouts including transparent backgrounds, and staying aligned with your intended visual style

On the safety side, Images 2.5 builds on OpenAI's existing safeguards, with checks on prompts and images to help prevent harmful outputs. It continues to use C2PA metadata and invisible watermarking to help identify images made with these tools, and OpenAI published more details in a system card.

Images 2.5 is rolling out today to ChatGPT, ChatGPT Work, and Codex users across all tiers, on desktop, mobile, and web.

Wrap-up

• ChatGPT Images 2.5 is a new image model with sharper details, more precise editing, and faster generation
• Generation latency is down by up to 50% compared with Images 2.0
• ChatGPT gets new creative controls: Sketch, templates, on-image comments, and shareable prompts
• Developers get two new API models, GPT-Image-2.5 Flare and GPT-Image-2.5 Sunburst
• Reference-photo fidelity and multi-turn editing consistency have both been significantly improved

This update is perfect for anyone who creates images in ChatGPT or wants to turn sketches and ideas into finished visuals, as well as developers looking to build image generation into their own products!