Google launches Nano Banana 2.1 with sharper visuals and better subject consistency
Google released Nano Banana 2.1, sharpening image realism and subject consistency as it rolls out across Gemini and Ads.
Google described the release in a social post and set out a fuller list of changes on its Gemini developers' website. Across 1K, 2K and 4K output resolutions, with 1K remaining the default, the model is said to deliver improved visual quality and realism. The same documentation says the update fixes tiling artifacts that appeared on wide and panoramic aspect ratios of 1:4, 4:1, 1:8 and 8:1 at 2K and 4K resolutions.
Text rendering and the layout of infographics are also described as more accurate. The model supports multi-image fusion with as many as 14 reference images, holding character consistency for up to four characters and object fidelity for up to ten objects. Google says Nano Banana 2.1 can be grounded using Google Web Search and Google Image Search, and that its Thinking levels are configurable at minimal, medium and high, with medium as the default.
Nano Banana 2.0 arrived in February and allowed the model to generate images at up to 4K resolution. Google describes the new release as a more efficient counterpart to Nano Banana Pro.
The rollout covers the Gemini app, AI Mode, Google AI Studio, Google Flow, Stitch by Google, Google Ads and the Gemini Enterprise Platform.