OpenAI launches ChatGPT Images 2.5 with faster generation

Investing.com -- OpenAI released ChatGPT Images 2.5 on Tuesday, introducing improvements to image quality, editing precision, and generation speed for its image creation tools.

The company said users generate more than 3 billion images weekly across ChatGPT Images and GPT-Image models in the API. The new version reduces image generation time by up to 50% compared with Images 2.0.

Images 2.5 produces more natural lighting and richer textures, according to OpenAI. The model better preserves subjects in reference photos and follows editing instructions more reliably across multiple turns.

OpenAI introduced several new features for ChatGPT users. Sketch allows users to draw directly in ChatGPT as a reference for final images. Templates provide starting points for popular image formats including flyers and product photos. Users can now place comments directly on images for focused editing and share prompts with others.

The update is available to all ChatGPT, ChatGPT Work, and Codex users across desktop, mobile, and web platforms.

For developers, OpenAI released two new models in the API. GPT-Image-2.5 Flare delivers the same quality, editing, and speed improvements. GPT-Image-2.5 Sunburst offers additional precision for detailed creative work with longer generation times.

The new models improve reference photo handling, allowing image subjects to appear more recognizable while maintaining distinctive features across different settings and visual styles.

Images 2.5 performs better at making focused edits while keeping other details unchanged, even with complex subjects and backgrounds. The model maintains consistency across multiple edits during longer conversations without degrading image quality over time.

The update includes better understanding of complex visual instructions and more accurate handling of real-world information. The model can manage more complex layouts, including transparent backgrounds, and better reflects visual styles.

OpenAI said the release builds on existing safeguards with checks on prompts and images to prevent harmful outputs. The company continues using C2PA metadata and invisible watermarking to identify images created with its tools.

You May Also Be Interested In

• Piper Sandler Starts Oklo (OKLO) at Overweight
• Raleigh Author Ruth Danoff's Debut Memoir Holy ADHD Hits #1 Amazon Best Seller Days After Launch
• Merck Veteran, Psychedelics Developer Tapped For Top FDA Job - Bloomberg