Google · none caching
Nano Banana Pro
Nano Banana Pro on Zumik: live pricing, context, and caching, routable by id or alias through one OpenAI-compatible endpoint.
Specifications
At a glance.
| Provider | |
| Family | gemini |
| Released | — |
| License | Proprietary |
| Context window | 131K tokens |
| Max output | 33K tokens |
| Modalities | text, image |
| Tool calling | Yes |
| Reasoning mode | No |
| Caching | none |
| Batch discount | 50% off |
Measured by Zumik
What reuse looks like here.
Pricing, context, and capabilities for Nano Banana Pro are live, but it is outside the flagship set Zumik benchmarks in depth, so measured reuse, capture, and warm TTFT are not shown yet. Run a workload estimate or route it by id to start collecting traces.
Call it
Same OpenAI client, this model.
from openai import OpenAI
client = OpenAI(base_url="https://api.zumik.ai/v1", api_key="zk_live_...")
r = client.responses.create(
model="nano-banana-pro-preview",
input="Draft a fix for the failing test.",
)
print(r.usage.input_tokens_cached) # confirm reuseFrequently asked
Nano Banana Pro, answered.
How much does Nano Banana Pro cost?
Nano Banana Pro is an open-weights model routed through Google Gemini. It is priced on the host's serverless size tier rather than a single published per-token list price, so it shows "—" here until profiled.
What is Nano Banana Pro's context window?
Nano Banana Pro supports a 131K-token context window with up to 33K output tokens.
Does Nano Banana Pro support prompt caching?
Yes. Google Gemini uses Implicit context caching caching. In the Zumik corpus, Nano Banana Pro shows a median cache capture of 75% on agent workloads.
Run Nano Banana Pro with reuse measured.
Point an OpenAI client at Zumik and see exactly how much of this model's input you are reusing.