The mysterious “Nano-Pana” project has been revealed to be the latest Google photo editor

Google has just upgraded the artificial intelligence image model, and it actually appears to be a very important step. The company has launched Gemini 2.5 Flash Image today, a large update that generates more intelligent and more flexible images.

The model for users allows the issuance of natural language claims not only to create images but also to combine current images and make more accurate adjustments without creating strange abnormalities. It also clicks on the “global knowledge” of Gemini to understand what it generates better.

This upgrade comes at a time when Google tries to bridge the gap with the leader of the industry, Openai. In the past, the photo was a major driver for Amnesty International. The use of Chatgpt increased in March when the company launched the original GPT-4O photo generator. The viral memes resulted in GHibli Studio resulting from the company GPU models meltedAccording to the chief executive, Openai, Sam Al -Tamman. Chatgpt is currently finished 700 million weekly users. In comparison, Google Sundar Pichai CEO revealed the company’s profit call in July -made by Gemini 450 million monthly usersIt is still lagging behind ChatGPT.

https://www.youtube.com/watch?

With his latest update, Google says that one of the largest headaches has been resolved in artificial intelligence. To date, maintaining letters or objects has been consistent with multiple adjustments represented a major challenge to artificial intelligence generators.

“You can now put the same crafts in different environments, display one product of multiple angles in new settings, or create consistent nude assets, all while maintaining the topic. ” The company wrote in a Blog post.

Google says users can now make very specific adjustments with only demand. For example, users can obliterate the image wallpaper, remove a stain from the T -shirt, change a person’s position, or even add a color to a black and white image.

Even before its official launch, the new model was running heads on the LMARNA assessment platform, where anonymous appeared under the name “Nano-Banana”. one X user Share how they used Nano-Paniana to change the Taman shirt in a picture. The result was amazingly good. Today, Google moved forward and claimed the ownership of the model, and revealed that Nano-Banana was actually a Gueini 2.5 Flash image.

In addition to being available on the Gemini app, the new model of developers can now be accessed through Gemini API, Google Ai Studio and Vertex Ai. Google has already built many template applications that use the new model on Google Ai Studio, and AI’s assistant coding the company, and he said to users they could be a symbol above it.

The company also said that some developers have already tried the application to see how it would be useful in real world scenarios, such as creating real estate menu cards, employee badges, and product sofas.