Google has introduced a new AI model, Nano Banana 2 Lite, for image generation. According to developers, the neural network is the fastest in its lineup. This was reported by Qazaqyia.kz citing Kursiv Media.
In a Google blog post, the company announced the creation of a lightweight version of the image generator, focusing on speed and low cost. The key feature of the 2 Lite model is its efficiency. According to the company, the neural network generates an image from a text prompt in an average of four seconds.
"Whether your workflow requires generating thousands of images or editing complex videos, you now have two new tools that allow you to work faster, make flexible changes, and bring creative ideas to life," the developers write.
Nano Banana 2 Lite is available not only to corporate clients. The model is already free in the Gemini chatbot. Simply log into your Google account and select the 3.1 Flash-Lite model. A Gemini Advanced subscription is not required.
According to the team, the model correctly understands text prompts, can add text inside images, and maintains a consistent appearance of characters during regeneration. Kursiv LifeStyle tested the neural network and asked it to create a panorama of Almaty.
Here is the result obtained in 5.9 seconds.
Earlier, Kursiv LifeStyle wrote about the Gemini 3.5 Live Translate neural translator. The new model translates over 70 languages in real time, including Kazakh and Russian.