Google has launched Nano Banana 2.1, a new image model built on Gemini 3.6 Flash that promises better image editing and lower API costs for developers.
The model follows Nano Banana 2, released in February, and is rolling out gradually across the Gemini app, AI Mode in Google Search, Google Ads and developer tools including AI Studio, Flow and Stitch.
What Nano Banana 2.1 Adds
Nano Banana 2.1 supports mask-based editing, where users select part of an image and only that area changes. It accepts up to 14 reference images at once, produces output up to 4K resolution and can check Google Search before showing real-world subjects.
On Gemini API pricing, a 1K image costs about 3.4 cents, down from 6.7 cents, while a 4K image costs about 7.6 cents, down from 15.1 cents. Batch jobs are 50% cheaper, although input costs are higher than before.
Google says the model scored 1,050 Elo in human preference tests, against 990 for Nano Banana 2. Independent tests are not yet available, and Google’s own model card notes that small text can look blurry at 1K resolution.
Source: ARY News
Read More: More tech and telecom news
