OpenAI Unveils GPT-Image 1.5 with Speedier Edits to Challenge Gemini's Nano Banana Pro

Advancements in AI Image Generation

In 2025, AI image generation models experienced significant progress, and OpenAI is continuing this trend as the year draws to a close. On Tuesday, the company unveiled GPT-Image Model 1.5, its latest image generator, which promises enhanced performance, improved instruction-following capabilities, and more accurate editing tools.

This new model is immediately available to all ChatGPT users globally and can also be accessed through OpenAI’s API under the name GPT-Image-1.5. Additionally, OpenAI introduced a dedicated image creation space within ChatGPT’s sidebar, featuring preset styles and trending designs. This update represents a shift from simple chat-based image prompts to a more visual-first workflow.

Fidji Simo, CEO of applications at OpenAI, wrote in a blog post that many people's first experience with ChatGPT involves transforming a text prompt into an image. “It’s a magical way to see what this technology can do, but the chat interface wasn’t originally designed for this,” she stated.

Enhanced Performance and Control

OpenAI claims that GPT-Image 1.5 runs up to four times faster than its predecessor. The model also follows instructions more closely, particularly during iterative edits. Users can now add or remove objects, change visual styles, adjust clothing, and refine specific areas without regenerating the entire image. This level of control addresses a long-standing issue in generative image tools, which often rework entire scenes instead of making targeted changes.

“Creating and editing images is a different kind of task and deserves a space built for visuals,” Simo wrote. The model also produces more legible text inside images, a challenge that has long plagued image generators, including OpenAI’s earlier models. Initial test images suggest improvements, though real-world testing will determine how consistent the results remain.

A Creative Studio Shift

Beyond the model itself, OpenAI is changing how users interact with images in ChatGPT. The new sidebar entry point functions more like a creative studio, offering inspiration through trending prompts and preset filters. “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,” Simo wrote.

OpenAI plans to expand visual elements across ChatGPT more broadly. Search results may soon include more images with clear sources, which could help with tasks like checking sports scores or converting measurements. “When you’re creating, you should be able to see and shape the thing you’re making,” Simo wrote.

Intensifying Competition

The release of GPT-Image 1.5 comes amid growing competition with Google. OpenAI CEO Sam Altman reportedly declared a “code red” last month after Google’s Gemini 3 and Nano Banana Pro models topped the LMArena leaderboard. Google has maintained momentum even after OpenAI launched GPT-5.2 last week, which the company promoted as its most advanced model for developers and professionals.

OpenAI had reportedly planned to release its next image model in early January before accelerating the timeline. GPT-Image 1.5 also arrives shortly after OpenAI signed an agreement with Disney to bring more than 200 iconic characters to ChatGPT images and Sora videos. Disney characters will not appear in GPT-Image 1.5 yet, with that feature expected in early 2026.

As image and video tools move closer to production-ready systems, OpenAI appears focused on speed, consistency, and control. The company aims to bridge the gap between imagination and execution, one image at a time.