Skip to main content
Current modelGoogle

Nano Banana 2.1 benchmarks, pricing and specs

Nano Banana 2.1 is a Google model released on October 6, 2026 that takes text, image and video input and produces image and text. It costs $0.0336 per image. It scores 1328 Elo on Arena Text-to-Image, #4 of 9 current models (Arena, measured October 7, 2026).

Per image
$0.0336
Arena Text-to-Image
1328 Elo · #4/9

Data verified October 12, 2026 · Dataset 2026-10-12

Share this

Key facts

Provider
Google
Model family
Nano Banana
Release date
October 6, 2026
Status
Current
Access
API only
API model id
gemini-nano-banana-2.1
Context window
Not reported
Max output
Not reported
Input
text, image, video
Output
image, text
Task families
image, multimodal
Declared capabilities
image generation, image editing, image to image

Nano Banana 2.1 pricing

It is the #2 cheapest of 6 current image models priced per image.

UnitPrice (USD)TierAppliesSource
1M input tokensNot reportedstandardCurrent
1M output tokensNot reportedstandardCurrent
1M input tokensNot reportedcachedCurrent
1M output tokens$30standard image outputCurrent
image$0.0336standard 1kCurrent
image$0.0504standard 2kCurrent
image$0.113standard 4kCurrent

Independent benchmark results

Scores come from independent evaluators that ran the model themselves, pinned to a dated snapshot. Rank is among current models in the same snapshot. Google's own launch figures are listed under strengths with the publisher named, because they are not measured the same way.

BenchmarkScoreRankMeasuredSource
Arena Text-to-Image1328 Elo#4 of 9October 7, 2026Arena · 2026-10-12
Arena Image Edit1428 Elo#4 of 8October 6, 2026Arena · 2026-10-12

Strengths, cautions and best uses

Strengths

  • An update to Nano Banana 2 (Gemini 3.1 Flash Image) with better visual quality, prompt adherence, multi-turn character consistency and text rendering at Flash-tier speed
  • Generates and edits at 1K, 2K and 4K, accepts text, image, video and PDF input, and fuses up to 14 reference images
  • At $0.0336 per 1K image it costs about half of its predecessor's $0.067

Cautions

  • Google prices output by image tokens; the per-image figures are Google's own equivalents and batch is half price
  • No independent image-quality benchmark is recorded here

Best for

  • High-volume product, marketing and social imagery where cost per image matters
  • Iterative editing that must keep characters and objects consistent across turns

Alternatives to Nano Banana 2.1

The closest current imagemodels by price. The cheapest model that scores as well or better is listed first; failing that, the cheapest one within 3% of the benchmark's scale.

Compare Nano Banana 2.1 with

Nano Banana 2.1 FAQ

How much does Nano Banana 2.1 cost?

Nano Banana 2.1 costs $0.0336 per image at the standard tier. It is the #2 cheapest of 6 current image models priced per image. Source: Google, verified October 11, 2026.

How does Nano Banana 2.1 score on independent benchmarks?

Nano Banana 2.1 scores 1328 Elo on Arena Text-to-Image, #4 of 9 current models (Arena, measured October 7, 2026) and 1428 Elo on Arena Image Edit, #4 of 8 current models (Arena, measured October 6, 2026). These are independent evaluations, not Google's own launch figures.

Is Nano Banana 2.1 open source or API only?

Nano Banana 2.1 is available through Google's API only; its weights are not published.

What are the best alternatives to Nano Banana 2.1?

The closest current alternatives by price and measured performance are Gemini 3.1 Flash Lite Image (same price per image, $0.0336), FLUX1.1 [pro] (19% more per image, $0.04 vs $0.0336), Gemini 3.1 Flash Image (34% more per image, $0.045 vs $0.0336) and grok-imagine-image (40% cheaper per image, $0.02 vs $0.0336).

Sources

Cite this page

Van Data Team. "Nano Banana 2.1 Benchmarks, Pricing and Specs." AI Model Benchmark Dataset 2026-10-12. https://vandatateam.com/ai-model-benchmarks/nano-banana-2-1

Reuse the data: JSON · CSV · or explore every model in the AI model benchmark comparison.

Production evaluation

Deciding whether Nano Banana 2.1 fits your workload?

Public benchmarks narrow the shortlist. We test the finalists on your own tasks, data and cost limits, then build the routing and review controls to run the winner in production.

  • Evaluation cases and acceptance thresholds tied to real work
  • Model routing, fallbacks and cost controls