Google cuts image costs with Nano Banana 2.1

Google has released Nano Banana 2.1 for image generation and editing, replacing Nano Banana 2 from February 2026. The new model lowers prices roughly by half while improving quality, text rendering, consistency, panoramas, and infographics.

Google cuts image costs with Nano Banana 2.1

Google has introduced Nano Banana 2.1, a new image generation and editing model that replaces Nano Banana 2, which was released in February 2026. The update is designed to make image creation cheaper while improving several areas that matter in practical production workflows.

The release does not appear to displace Nano Banana Pro as Google's top option for image quality. Instead, Nano Banana 2.1 sits as a faster and less expensive counterpart, using Gemini 3.6 Flash while Nano Banana Pro uses the most recent Pro model, 3.1.

Lower Prices Change The Calculation

The clearest shift is cost. Google is cutting prices roughly in half compared with Nano Banana 2, which makes Nano Banana 2.1 more attractive for users who need many generated or edited images rather than a small number of premium outputs.

A 1K image now costs 3.36 cents, down from 6.70 cents. A 4K image now costs 7.56 cents, down from 15.10 cents. Nano Banana Pro remains much more expensive at 13.40 cents per 1K image.

That pricing creates a more distinct choice between efficiency and maximum output quality. For high-volume work, Nano Banana 2.1 offers a lower-cost path. For users who care most about the strongest possible result, Nano Banana Pro likely remains the more capable option in practice.

What Nano Banana 2.1 Improves

Google says Nano Banana 2.1 improves on previous versions across the board. The specific areas named include visual quality, text rendering, character consistency across conversation turns, wide panoramas, and infographics.

Those improvements matter because image models are often judged not only by how polished an individual output looks, but by whether they can maintain details across edits and follow-up prompts. A model that can preserve characters and objects more reliably is more useful for iterative creative work.

Nano Banana 2.1 can process up to 14 reference images at once. It can keep up to four characters and ten objects consistent, which gives users more room to guide an image with source material while asking the model to preserve important subjects.

The model also connects to Google Search. It offers minimal, medium, and high thinking levels, and those levels affect image quality.

Why Pro Still Matters

Nano Banana 2.1 may outperform Pro by a wide margin in some benchmarks, according to the source material. But that comparison comes with an important caveat: Nano Banana 2 also matched Pro in those tests, while real-world results often favored Nano Banana Pro.

In practice, Nano Banana Pro often still produces noticeably better images. That means benchmark results alone may not settle the choice for users comparing Google's image models.

Google describes Nano Banana 2.1 as the more efficient counterpart to Pro. That framing is useful because it sets expectations: this is not simply a replacement for the premium model, but a lower-cost, Flash-tier option meant to deliver stronger results than the prior non-Pro version.

The next major gain in image quality could come with Google's announced Pro model, Gemini 4 Argon. For now, Nano Banana 2.1 appears to be the more economical update, while Pro remains the likely quality leader.

Where It Is Rolling Out

Nano Banana 2.1 runs at Flash-tier speed and supports 1K, 2K, and 4K resolutions. It is already rolling out across Google's products, which gives the model a broad path into both consumer and work-oriented image tools.

The rollout includes the Gemini app, AI Mode in Google Search, Google AI Studio, Flow by Google, Stitch by Google, Google Ads, and the Gemini Enterprise Platform.

For users already working inside Google's AI products, that means Nano Banana 2.1 may become available through tools they already use rather than as a separate destination. The model's lower pricing and wider availability could make it a default choice for everyday generation and editing tasks where Pro-level output is not required.

The Previous Model Is Going Away

Google will shut down the previous model, gemini-3.1-flash-image, on October 29, 2026. More details are available in the model card.

That date gives users a clear endpoint for the older model. Anyone relying on the previous version will need to account for the change before the shutdown, especially if their workflows depend on consistent behavior from the older image model.

The broader message is straightforward: Nano Banana 2.1 is a cheaper and improved Flash-tier image model, while Nano Banana Pro remains the stronger quality option for many practical uses. The best choice depends on whether cost, speed, or top-end image quality matters most for the work at hand.