Google Boosts Gemini AI Image Capabilities in Latest Salvo Against ChatGPT

Google introduced Gemini 2.5 Flash Image on Tuesday, providing a brand-new AI design that creates and modifies images with more accuracy and character consistency than previous tools– trying to close the space with OpenAI’s ChatGPT.

The tech giant’s push to incorporate sophisticated image modifying into Gemini shows a more comprehensive push amongst AI platforms to consist of image generation as an essential function. The brand-new tool, now readily available throughout Gemini apps and platforms, lets users modify visuals utilizing natural language– dealing with intricate jobs like present modifications or multi-image blend without misshaping faces or scenes.

In an article, Google stated the design permits users to “position the very same character into various environments, [and] display a single item from numerous angles … all while maintaining the topic.”

The design initially appeared under the pseudonym “nano-banana” on crowdsourced screening website LMArena, where it drew attention for its smooth modifying. Google validated Tuesday it lagged the tool.

Google stated the system can fuse numerous images, preserve character consistency for storytelling or branding, and incorporate “world understanding” to analyze diagrams or integrate recommendation products– all within a single timely.

The design costs $30 per million output tokens– about 4 cents per image– on Google Cloud. It’s likewise being dispersed through OpenRouter and fal.ai.

OpenAI presented the GPT-4o design in Might 2024 and included image generation in March 2025, which assisted press ChatGPT’s use above 700 million weekly active users. Google reported 400 million month-to-month active Gemini users in August 2025, which would suggest weekly use that significantly routes OpenAI.

Google stated all outputs will consist of an unnoticeable SynthID watermark and metadata tag to mark them as AI-generated to resolve issues around abuse and credibility.