Key Points
- OpenAI releases GPT Image 1.5, promising up to 4x faster generation and better instruction following.
- New granular edit controls allow precise adjustments to facial expressions, lighting, and color tone.
- The model is accessible to all ChatGPT users and via API, expanding on the earlier GPT Image 1.
- Launch follows an internal “code red” memo highlighting competition with Google’s Gemini 3 and Nano Banana Pro.
- A redesigned sidebar interface works like a creative studio, offering post‑production features.
- Fidji Simo emphasizes adding more visual elements to search queries and everyday tasks.
New Model Details
OpenAI announced the release of GPT Image 1.5, the latest iteration of its ChatGPT image generation service. The update is described as delivering up to 4x faster image generation speeds while improving the model’s ability to follow detailed instructions. Users can now make precise edits—such as adjusting facial expressions or tweaking lighting—without the model reinterpreting the entire image, preserving visual consistency across changes.
Competitive Landscape
The launch comes amid an internal memo that labeled the situation a “code red,” reflecting OpenAI’s strategic response to Google’s recent advances. Google’s Gemini 3 and the Nano Banana Pro image generator have topped the LMArena leaderboard across multiple benchmarks, prompting OpenAI to accelerate its roadmap. Despite Google’s lead, OpenAI has also introduced GPT‑5.2, positioning both its language and visual models as direct competitors to Google’s offerings.
User Experience Enhancements
GPT Image 1.5 introduces a dedicated entry point in the ChatGPT sidebar that functions “more like a creative studio,” according to Fidji Simo, OpenAI’s CEO of applications. The new interface provides users with post‑production controls for facial likeness, lighting, composition, and color tone, allowing for granular edits that maintain visual continuity. Simo highlighted that the updated image viewing and editing screens make it easier to create images that match a user’s vision or draw inspiration from trending prompts and preset filters.
Strategic Context
The model is available to all ChatGPT users and via the API, expanding access beyond the earlier GPT Image 1 release, which debuted in April. OpenAI had originally planned a new image generator for early January, but the rollout was accelerated following the “code red” memo. The company’s broader ambition, as outlined by Simo, is to integrate more visual elements into search queries and other tasks—such as converting measurements or checking sports scores—so that visuals can tell a story more effectively than words alone.
Outlook
By positioning GPT Image 1.5 as a production‑ready tool with rapid generation and refined editing capabilities, OpenAI aims to close the gap with Google’s leading image generators while reinforcing its own ecosystem of AI services. The release signals a continued intensification of the AI race, with both firms pushing forward new models and features to capture market share.
Source: techcrunch.com