Google Cuts AI Image Generation Costs in Half With Nano Banana 2.1
The tech giant releases its latest text-to-image model, promising improved performance and developer tools at substantially lower prices.
Key takeaways
- Google has released Nano Banana 2.1, its latest text-to-image and photo-editing AI model.
- The new model cuts standard API pricing in half compared to its predecessor, charging $0.0336 per 1K image.
- Performance claims rely on Google's internal testing, showing notable gains in visual design and subject consistency.
Google rolled out Nano Banana 2.1 on Tuesday, introducing the latest iteration of its artificial intelligence model designed to generate and edit images from text prompts. The new software is expanding across the company's ecosystem, including the Gemini app, Google Search's AI Mode, Google Ads, and developer platforms such as Google AI Studio, Flow, and Stitch.
The original Nano Banana model previously propelled Gemini to significant adoption. In September 2025, a feature turning user selfies into collectible figurines drove Gemini to the top of major app store charts, temporarily unseating competing applications. Version 2 followed in February, built on the Gemini 3.1 Flash Image architecture and incorporating live Google Search data to improve the accuracy of real-world depictions.
Upgrades and Technical Capabilities
Google highlights three primary enhancements in the version 2.1 update: refined visual design, sharper mask-based editing that restricts modifications to isolated regions, and improved subject consistency. The latter feature ensures that specific people or objects retain their appearance across multiple sequential edits.
In internal text-to-image preference evaluations measured via the ELO system, Nano Banana 2.1 achieved 1,050 points. This compares to 990 points for Nano Banana 2 and 935 points for Nano Banana Pro.
Technically, the model processes up to 14 reference images simultaneously, tracking up to four characters and ten objects. It generates outputs in resolutions up to 4K with aspect ratios as wide as 8:1. Developers can also adjust the processing duration before image generation and connect the model to Google Search and Google Image Search.
Pricing and Efficiency
Through Google's developer application programming interface, standard 1K images now cost $0.0336, marking a 50% reduction from Nano Banana 2's price of $0.067. Generating 4K images costs $0.0756 compared to the previous $0.151, while batch processing jobs receive an additional 50% discount.
Why it matters
As competition in generative artificial intelligence intensifies, cost reduction and efficiency have become central battlegrounds for major tech firms. By halving the price of API access while simultaneously claiming performance gains, Google is increasing the economic pressure on rival image generation providers. Lower costs lower the barrier to entry for developers and enterprises building applications on top of these models, potentially accelerating commercial adoption across advertising, design, and media industries.
What to watch
Observers will monitor independent benchmarks to verify Google's performance claims, as current metrics rely entirely on the company's internal testing. Market adoption rates among developers utilizing the new API pricing structure will also indicate whether these cost reductions successfully drive platform migration away from competing services.
For Information Purposes Only. Not Investment Advice. How we work