Google has unveiled Nano Banana 2.1, the latest iteration of its artificial intelligence model designed for image generation and manipulation.
This enhanced model boasts the ability to produce visually superior images, adhere more closely to user directives, and execute image modifications with greater efficacy compared to its predecessor.
Additionally, there’s a financial dimension: according to Google, Nano Banana 2.1 generates images at nearly half the cost of the earlier version.
The new model is built upon Google’s Gemini 3.6 Flash architecture and is being integrated across a variety of the company’s platforms, encompassing the Gemini app, Google AI Studio, the Gemini API, the AI component of Google Search, Google Ads, Flow, and Stitch.
What Enhancements Does Nano Banana 2.1 Offer?
In line with its earlier variants, Nano Banana 2.1 permits users to generate images by merely articulating their desires in a text prompt.
Google asserts that the updated model demonstrates improved comprehension of intricate instructions and maintains essential elements’ consistency during image alterations.
Users have the capability to incorporate multiple reference images, allowing the model to synthesize them into a new creation.
Nano Banana 2.1 accommodates up to 14 reference images concurrently and can preserve consistency for up to four characters and ten objects.
This functionality is particularly advantageous for individuals looking to produce a cohesive series of images featuring the same subject, product, or object across diverse scenarios.
The model also excels in modifying existing images according to straightforward textual commands. Users can alter an individual’s attire, modify or substitute the background, append or remove objects, or impart a distinct appearance and style without the necessity of restarting the image creation process from scratch.
Moreover, Google states that Nano Banana 2.1 enhances the handling of text within AI-generated visuals, ensuring better readability in posters, diagrams, infographics, and other visual formats where clarity is paramount.
The model supports image generation across 1K, 2K, and 4K resolutions. Users can select varying levels of AI reasoning contingent on their desired use of processing power.
One remarkable innovation is the model’s integration with Google Search (Search grounding). Nano Banana 2.1 can leverage data from Google Search and Google Image Search, facilitating the creation of images grounded in real-world information.
Cost Analysis for Nano Banana 2.1’s Image Generation
The most notable alteration may well be the pricing structure. Generating a 1K image through the API now incurs a cost of approximately $0.0336, a marked reduction from the previous $0.067.
Meanwhile, a 4K image is priced around $0.0756, down from $0.151. Consequently, developers can now produce images for roughly half the cost of earlier iterations.
For developers and enterprises engaged in the mass production of images, this price discrepancy can lead to substantial savings over a brief period.
Therefore, Nano Banana 2.1 represents not merely an enhancement of Google’s imaging capabilities but also a transformative solution for economically viable large-scale image production.
The rollout of the model across Google’s product suite allows both developers and everyday users to leverage its advanced image generation and editing functionalities.
Source link: Thehansindia.com.
