Nano Banana Pro Pricing: What It Actually Costs Per Image
Most write-ups on Nano Banana Pro pricing repeat a wrong number: if you’ve seen Pro quoted at $0.15 or $0.30 an image, you’ve seen a figure copied across a dozen blogs without checking it. The real figure, from Google’s own pricing page, is $0.134 per image at 1K or 2K and $0.24 at 4K — and you can prove it from the token math, which we’ll show. Pro costs roughly three times the base Nano Banana model, there’s no free tier on the paid API, and a batch option quietly cuts the price in half.
This is the Pro-specific breakdown: the correct per-image ladder, why the popular numbers are off, whether Pro’s premium over base is worth it, and how to pay less. (For the whole Nano Banana family priced side by side, see the full Nano Banana price breakdown; here we go deep on Pro alone.) Every figure is from Google’s and the vendor’s own pages on July 22, 2026 — and since Pro is a preview model, treat them as a dated snapshot.
The short answer
Nano Banana Pro costs $0.134 per image at 1K/2K resolution and $0.24 at 4K, billed on Google’s paid API — with no free tier. It’s a token-metered model: image output is priced at $120 per million tokens, a 1K/2K image consumes 1,120 tokens, and a 4K image consumes 2,000. If you can batch your jobs, Google’s Batch mode halves both rates.
| Nano Banana Pro (per image) | Standard | Batch (−50%) |
|---|---|---|
| 1K / 2K (1,120 tokens) | $0.134 | $0.067 |
| 4K (2,000 tokens) | $0.24 | $0.12 |
Google’s Gemini API pricing, captured July 22, 2026 — gemini-3-pro-image at $120/1M output tokens, equivalent to $0.134 (1K/2K) and $0.24 (4K) per image, Free Tier “Not available”.
Scaled to a real workload, the arithmetic is simple. Here’s what 1,000 images a month costs on each path:
| 1,000 images/month (1K/2K) | Monthly cost |
|---|---|
| Standard API | $134 |
| Batch API (−50%) | $67 |
| Aggregator (Modellix, $0.1265) | $126.50 |
Pro is a preview model — prices can change, so confirm against the official page before you commit.
The rest of this guide shows the math, corrects the numbers you’ll see elsewhere, and helps you decide whether Pro is the right tier at all.
Is Nano Banana Pro free?
Short answer: not on the API. Google’s pricing page lists the Free Tier for gemini-3-pro-image as “Not available.” Every image you generate through the paid API is billed — there’s no free quota, which is the honest answer to the very common nano banana pro free search.
The consumer side is where it gets fuzzy, and worth stating carefully rather than guessing. Nano Banana Pro is available inside Google’s Gemini app to subscribers, and free-tier app users may get limited access before being routed to the lighter base model — but Google doesn’t publish a specific “free Pro image” allowance, so we won’t invent one. The reliable takeaway: if you’re building on the API, budget for every image; if you’re a casual user in the Gemini app, any Pro access you get is limited and undocumented, not a dependable free source. Assume you pay per image.
What Nano Banana Pro costs per image
Here’s the part most write-ups get wrong. Pro isn’t priced per image directly — it’s priced per token, and images are converted to tokens at a fixed rate. That’s what makes the real number verifiable instead of a guess:
- A 1K or 2K image = 1,120 tokens → 1,120 × ($120 ÷ 1,000,000) = $0.1344, i.e. $0.134.
- A 4K image = 2,000 tokens → 2,000 × ($120 ÷ 1,000,000) = $0.24.
That derivation is why the widely-copied $0.15 and $0.30 figures are wrong: they don’t reconcile with Google’s published token count and per-token rate. If a page can’t show the token math, treat its Nano Banana Pro price as a guess. (There’s also a tiny input cost — an uploaded reference image is ~560 tokens, about $0.0011 — negligible next to output.)
One practical wrinkle: the sticker price is per generated image, not per usable one. If, say, four in five generations meet your bar — an illustrative ratio, not a measured figure — your real cost per keeper is roughly $0.134 ÷ 0.8 ≈ $0.17 at 1K/2K. It’s not a fee Google charges; it’s the number to budget with, and it applies to every image model, not just Pro.
Is Pro worth 3× the base model?
A quick naming trap before the numbers, because it trips people up in any nano banana pro vs nano banana comparison: Pro is built on Gemini 3 Pro (gemini-3-pro-image), while the mid-tier “Nano Banana 2” is Gemini 3.1 Flash and the base model is Gemini 2.5 Flash. The version numbers don’t rank the models — Pro’s “3” is the higher-end tier despite the smaller number.
The base Nano Banana model (gemini-2.5-flash-image) runs about $0.039 per image, so Pro’s $0.134 is roughly 3.4× more. Whether that premium is worth it comes down to what Pro actually buys you: higher fidelity, native 4K output, stronger prompt adherence, and better handling of text and fine detail. If the image is the deliverable — a hero shot, a print-resolution asset, dense text rendered cleanly — Pro earns its price. If you’re generating thumbnails, drafts, or high-volume variations where “good enough” wins, base is nearly a quarter of the cost and the better call.
The useful mental model: don’t default to Pro because it’s the flagship. Match the tier to the job. Many pipelines use base for iteration and reserve Pro for the final render — paying the 3.4× premium only on the images that ship.
How to pay less
Use Batch mode. Google’s Batch (asynchronous) pricing is half the standard rate — $0.067 per 1K/2K image and $0.12 at 4K. If your workload doesn’t need images back in real time — overnight renders, bulk generation, background jobs — this is a straight 50% cut with no quality change. It’s the single biggest lever on Pro’s cost and the one most roundups skip.
Don’t pay for 4K you won’t use. The jump from 2K to 4K nearly doubles the price ($0.134 → $0.24). Unless you’re delivering at print or large-display resolution, 2K is the same $0.134 as 1K and covers most screen use.
Budget by usable image, not generated image. As above, factor your acceptance rate into the number you plan around — it’s the honest cost, and it makes Batch’s savings matter even more at volume.
Nano Banana Pro pricing: Google direct vs an aggregator
You can reach Pro two ways, and the cost differs at the margin.
Direct from Google — via the Gemini API — you get the rates above, including Batch’s half-price mode, billed to a Google Cloud account. If you’re already there, it’s the natural path and, with Batch, the cheapest. The same model is also offered through Vertex AI for enterprise setups, where add-ons like search grounding can carry separate charges — worth checking the Vertex pricing if that’s your environment.
Through an aggregation layer, you call the same model behind one pay-as-you-go key alongside other providers, which helps if you’re comparing models or want a single bill and per-call cost visibility. Modellix, for example, carries Nano Banana Pro at $0.1265 per image for 1K/2K and $0.2093 at 4K, with an edit endpoint and per-call logging that records the cost and latency of each generation — the same key that reaches Google’s other image models if you want to call a text-to-image API across providers from one place.
Modellix’s Nano Banana Pro pricing, captured July 22, 2026 — $0.1265 (1K/2K) and $0.2093 (4K) per image.
Two honest caveats. Its standard rates sit just under Google’s standard rates — but Google’s Batch mode ($0.067 / $0.12) is lower than the aggregator’s price, so if you can batch and you’re on Google Cloud, direct is cheaper; we’re not claiming the aggregator is the cheapest path, only the most convenient when you’re juggling providers or need synchronous calls. And Nano Banana Pro is a Google model (gemini-3-pro-image); an aggregator distributes it, it doesn’t build it.
FAQ
How much does Nano Banana Pro cost?
$0.134 per image at 1K/2K and $0.24 at 4K on Google’s paid API ($120 per million output tokens; 1,120 tokens per 1K/2K image). Google’s Batch mode halves that to $0.067 and $0.12.
Why do some sites say $0.15 or $0.30?
Those figures don’t match Google’s published token math (1,120 tokens × $120/1M = $0.134). They’re rounded or mistaken — the token-derived $0.134 / $0.24 is the correct pair.
Is Nano Banana Pro free?
Not on the API — its Free Tier is listed as “Not available.” Some limited access exists in Google’s consumer Gemini app, but Google doesn’t document a specific free Pro allowance, so don’t rely on one for production.
Does Nano Banana Pro have a subscription?
No — the API is pay-per-image (by tokens), not a subscription. Any monthly fee you see refers to Google’s consumer Gemini plans, which are a different product from the API you’d build on.
Is Nano Banana Pro worth it over the base model?
It’s ~3.4× the base model’s ~$0.039/image. Worth it when you need 4K, top fidelity, or clean text and fine detail; overkill for drafts and high-volume thumbnails, where base is the cheaper, sensible choice.
