OpenAI has introduced its new image generation model, gpt-image-1.5. Offering faster performance, superior editing, and lower costs, it arrives as a direct competitor to Google’s Nano Banana models. In March, OpenAI launched a new model that fundamentally enhanced ChatGPT’s image generation capabilities.
This update went viral almost instantly, bringing millions of new users to ChatGPT; in fact, users generated over 700 million images in just one week. By April, the same technology was made available to developers via the gpt-image-1 API.
Google’s “Nano Banana” Move
In September, Google introduced its new image generation model, codenamed Nano Banana. The model quickly gained significant traction and substantially increased the usage of the Gemini platform. Surpassing gpt-image-1 in both creation and image editing, Nano Banana was rapidly adopted by developers worldwide.
Google sustained this momentum last month with the announcement of Gemini 3 Pro Image (Nano Banana Pro). The new model aims to produce the most accurate images to date by leveraging Gemini’s advanced reasoning capabilities and real-world knowledge.
OpenAI’s Counter-Attack: gpt-image-1.5
Today, OpenAI announced gpt-image-1.5, a new image generation model designed to compete more directly with Google’s Nano Banana series. The new model stands out particularly for its visual editing precision.
Better Editing, Clearer Text
gpt-image-1.5 offers significant improvements in understanding and executing user instructions while preserving critical details such as logos and faces much more effectively. OpenAI has also enhanced the readability of text within generated images. A visible improvement has been achieved, especially in small and dense text elements.
Cheaper and More Efficient for Developers
The new model also provides advantages for developers. Compared to the previous version, gpt-image-1.5 is 20 percent cheaper in terms of image input and output costs. While costs remain manageable through quality settings, OpenAI states that a powerful performance is delivered even at lower quality levels.
Up to 4x Faster Experience for ChatGPT Users
For end-users, gpt-image-1.5 now directly powers the image generation infrastructure of ChatGPT. According to OpenAI, the new model can make accurate edits while maintaining details like lighting, composition, and personal appearance, and it operates up to 4 times faster than the previous model.
New “Images” Hub Exclusive to ChatGPT
OpenAI has also launched a new Images hub dedicated to ChatGPT. Accessible via the mobile app and chatgpt.com, this section allows for a quick start to image generation with ready-made filters and sample prompts. The company notes that these pre-set contents will be regularly updated according to current trends.
Gradual Rollout to All Users
The new image generation model and the ChatGPT Images experience are being rolled out gradually to all ChatGPT users.
You Might Also Like;
- How Google DeepMind’s GNoME Just Fast-Forwarded History by 800 Years
- The Growing Risk in Low Earth Orbit: Humanity’s Space Adventure Could End Abruptly
- Is Solar Powered Crypto Mining Profitable?
