← All models: AI Image API Cost Calculator

Nano Banana 2.1 Cost Calculator

Google's current workhorse image model (GA Oct 6, 2026) and the designated replacement for gemini-3.1-flash-image. Output billed per token: a 1K image is 1,120 output tokens at $30 per million. Input tokens tripled vs the predecessor ($1.50/1M), which matters for edit-heavy flows with many reference images.

Lowest rate
$0.017
per image · Batch API (−50%) · 1K
Default plan
$0.034
per image · Standard · 1K
10K images / month
$336.00
output fees, before retries
Per image
–
Per usable image
–
after retries
Monthly
–

Nano Banana 2.1 cost at volume (Standard, 1K)

Images / monthOutput fees (1 attempt each)At 2 attempts per usable image
1,000$33.60$67.20
10,000$336.00$672.00
100,000$3,360.00$6,720.00
1,000,000$33,600.00$67,200.00

Full Nano Banana 2.1 rate table

Plan1K2K4K
Standard$0.034$0.050$0.076
Batch API (−50%)$0.017$0.025$0.038

API model ID: gemini-nano-banana-2.1 · Source: Gemini API pricing page (Google), October 2026. Rate source ↗ · Verified 2026-10-12

Nano Banana 2.1 pricing FAQ

How much does Nano Banana 2.1 cost per image?

Nano Banana 2.1 starts at $0.017 per image (Batch API (−50%), 1K) on published API list rates verified 2026-10-12. The default plan at 1K costs $0.034 per image before retries.

How much does Nano Banana 2.1 cost for 10,000 images a month?

At $0.034 per image (Standard, 1K), 10,000 generations cost $336.00 in output fees. Budget roughly double if you average two attempts per usable image — $672.00 — and more if your edits carry heavy reference-image input, which token-billed models charge separately.

Does Nano Banana 2.1 offer batch pricing?

Yes — the Batch API plan halves per-image output prices (for example $0.017 instead of $0.034 at 1K). For any volume that does not need interactive turnaround, batch is the single biggest saving available.

Is Nano Banana 2.1 billed per image or per token?

Nano Banana 2.1 bills per token under the hood; the per-image figures on this page are the vendor's own conversions at each size and quality tier. Input tokens (prompts and reference images) are charged on top of the output figures.

Comparing image models? The main image calculator ranks Nano Banana 2.1 against Nano Banana 2.1, GPT Image 2.5, FLUX and the rest at your exact volume.