AI-Altered 'Nano Bananas' from Gemini Offer Entertaining Editing, Yet Suffered Numerous Blunders
Google has unveiled a new generative AI model named Gemini 2.5 Flash Image, designed to revolutionise photo editing. This latest addition to the Gemini family, affectionately referred to as the "nano bananas" model by AI enthusiasts, is now available for free, with paid plans starting at $20 per month for more models and higher usage caps.
The Gemini 2.5 Flash Image model offers a unique blend of speed and creativity. It can produce completed images quickly, usually under 15 seconds, making it an ideal choice for those who want to make big edits quickly. However, it may not be suitable for those seeking more precise tools, as the model automatically generates square images and has limitations in adapting images to other dimensions.
One of the key features of Gemini 2.5 Flash Image is its ability to add elements to existing images and blend them seamlessly. This makes it a powerful tool for creators, who have already used Google's AI to generate over 100 million AI videos with the AI filmmaker tool, Flow.
However, the model does have some drawbacks. Resolution issues have been reported, with many photos experiencing a reduction in quality after going through the Gemini model. This may annoy photographers who value high-quality images. Google is aware of these issues and is actively working to improve the resolution and dimensions in the Gemini nano bananas model.
Google has also released updates for other AI models this year. The AI video generator Veo 3 was updated and stunned with synchronized audio, while Adobe Express and Firefly users can now access the new Gemini model. The model can be accessed through Google AI Studio for paying subscribers.
It's important to note that Google's AI prohibitive use policy outlaws the creation of illegal or abusive material. Users are encouraged to use the tools responsibly and adhere to the policy.
In terms of privacy, Google's Gemini privacy policy allows the use of uploaded information for improving AI products. Creators can rest assured that each prompt uses anywhere from one to two thousand tokens, depending on the level of detail required, ensuring a high degree of privacy and security.
In conclusion, Gemini 2.5 Flash Image offers a unique blend of speed, creativity, and capabilities, making it an exciting addition to the world of AI photo editing. While it may not be perfect, it's clear that Google is committed to investing heavily in its generative media models and improving them over time.
