Google released the Nano Banana 2.1 image model, and the developer called it about half the price of the previous generation.

ChainCatcher news, Google released the latest image generation and editing model Nano Banana 2.1 on October 6, and has been launched in the Gemini app, AI mode of Google search, Google Ads, Google AI Studio, Flow, Stitch and other developer tools. The model generates images from textual descriptions or edits to existing photos.

According to Google, this upgrade focuses on three areas: visual design, mask-based editing (marking an area and only that area changes), and subject consistency, that is, the person or object retains its original characteristics after multiple edits. In the Vincentian overall preference test, Nano Banana 2.1 scored 1050 ELO points, up from 990 points for Nano Banana 2 and 935 points for Nano Banana Pro. The model can process up to 14 reference images at the same time, track 4 characters and 10 objects, output up to 4K, with a maximum aspect ratio of 8: 1. Developers can also set the thinking time before its generation, and let it be fact-checked based on Google Search and Google Image Search.

When invoked through the Google Developer API, standard 1K images are priced at $0.0336, about half of the $0.067 for Nano Banana 2; 4K images are $0.0756, compared to $0.151 for the previous generation, and 50% off bulk tasks. The above performance data comes from Google's own tests.