Google officially releases Nano Banana 2.1, with image-output prices falling by approximately half

Article

Google Nano Banana 2.1 released: performance improved,API but the price was cut in half

Beating AI News reports that Google has officially released Nano Banana 2.1, an upgraded version of Nano Banana 2, still positioned as a highly efficient image-generation and editing model. The main improvements focus on image quality, instruction following, text generation, and multi-turn character consistency.Gemini、AI Mode、AI Studio、Flow、Stitch、Google Ads and other products have already begun rolling it out,API with the model designation gemini-nano-banana-2.1.

Image editing is a major focus this time. 2.1 supports 1K、2K and 4K, can use up to 14 reference images simultaneously, and maintain consistency for up to 4 characters and 10 objects. Google has also enhanced mask editing—changing only specified areas—and improved infographic layout, text in images, and ultra-wide image generation. The model can now select among three reasoning-intensity levels: minimal, medium, and high.

In Google’s own model card, with thinking enabled, 2.1 scores higher in point estimates than Nano Banana 10 and Nano Banana Pro across all 2 listed evaluations, including text-to-image generation, general editing, multi-person consistency, and mask editing. The independent Arena leaderboard also confirms a clear improvement: 2.1 currently ranks 5 in text-to-image generation and 6 in single-image editing, the highest among Google models, but it still trails GPT Image 2.5 and GPT Image 2.

The price, however, has fallen. For standard API, the prices for 1K、2K、4K image outputs are 0.0336、0.0504 and 0.0756 USD respectively—approximately half the price of Nano Banana 2. However, the input price has risen from token 0.50 USD per million to 1.50 USD, while text and thinking output prices have also risen from 3 USD to 7.50 USD. The more reference images used and the heavier the reasoning, the smaller the cost advantage from the lower image-output price.

Original link https://m.theblockbeats.info/flash/370524

Key points

The report says Google has officially released Nano Banana 2.1, an upgraded version of Nano Banana 2.
The report says Nano Banana 2.1 supports 1K、2K and 4K, can use up to 14 reference images simultaneously, and maintain consistency for up to 4 characters and 10 objects.
The report says Gemini、AI Mode、AI Studio、Flow、Stitch、Google Ads and other products have begun rolling out Nano Banana 2.1。
The report says that with thinking enabled, Nano Banana 2.1 has higher point estimates in all 10 listed evaluations than Nano Banana 2 and Nano Banana Pro.
The report says Nano Banana 2.1 currently ranks 5 in text-to-image generation and 6。
The standard prices listed in the report for API image outputs are: 1K 0.0336 USD、2K 0.0504 USD、4K 0.0756 USD。
The report says the input price rose from token 0.50 USD to 1.50 USD, while text and thinking output prices rose from 3 USD to 7.50 USD。

AI-assisted interpretation

The following is analysis, separate from reported facts. Verify important claims independently.

The core of this report is that Nano Banana 2.1 could potentially improve image generation and editing quality while lowering the per-image output price; however, input, text output, and thinking output prices have increased, so total costs are not necessarily lower when there are more reference images or the reasoning intensity is higher.

Why it matters to readers

For developers using image generation API, it is not enough to look only at the image-output price; actual costs must also account for input volume, the number of reference images, and thinking output volume.

If you generate images only occasionally, focus on the output price; if you batch-edit, use multiple reference images, or enable higher reasoning intensity, pay attention to the complete invocation cost.

Risks and unknowns

The original is a secondary report and does not provide Google’s official model documentation, pricing page, or product page as primary evidence.
The report’s product rollout scope,API model designation, evaluation results, and prices are not confirmed by the evidence to apply to all regions and accounts.
The evidence does not provide the actual proportions of input, text output, and thinking output for typical tasks, so it cannot confirm that overall costs will necessarily decrease.

Google officially releases Nano Banana 2.1, with image-output prices falling by approximately half

Article

AI-assisted interpretation

Why it matters to readers

Risks and unknowns

Related Developments

Related concepts

Keep following

Security Center