OpenAI Releases New Image Generation Model
OpenAI announced on December 16 the release of “GPT Image 1.5,” a new image generation model for ChatGPT. The model delivers better instruction-following, more precise editing, and up to 4x faster image generation speeds.
Counter-Move Against Google
This announcement follows CEO Sam Altman’s internal “code red” declaration last month. After Google released “Gemini 3” and the viral image generation model “Nano Banana Pro,” which topped the LMArena leaderboard across multiple benchmarks, OpenAI is intensifying its offensive to reclaim market share.
Following last week’s GPT-5.2 release, the company has accelerated the launch of its new image generation model, which was originally planned for early January.
Key New Features
1. Improved Editing Precision
- Edit while maintaining consistency in facial likeness, lighting, composition, and color tone
- Specific changes like “adjust the facial expression” or “make lighting colder” are applied without reinterpreting the entire image
2. 4x Faster Generation Speed
- Up to 4x faster compared to previous models
3. Creative Studio Feature
- New dedicated image area in the ChatGPT sidebar
- Inspiration through trending prompts and preset filters
Comment from OpenAI Applications CEO Fidji Simo
Fidji Simo, OpenAI’s CEO of Applications, wrote in an official blog post: “The new image viewing and editing screens make it easier to create images that match your vision or get inspiration from trending prompts and preset filters.”
Available to All ChatGPT Users
GPT Image 1.5 is available starting December 16 to all ChatGPT users and via the API. OpenAI’s previous image model release, “GPT Image 1,” was in April this year.