llm-benchmarks
githubdrose.io

Nano Banana 2 (Gemini 3.1 Flash Image) Benchmarks


Measured on google

4 runs
google site
Avg Tokens / Second
55.90
Decode Tokens / Second
Avg Time to First Token (ms)
1180.00
Runs Analysed
4
Last Updated
Sep 1, 2026, 03:31 AM

Throughput distribution

Throughput over time

Nano Banana 2 (Gemini 3.1 Flash Image)google54tok/s

Generated throughput, including reasoning tokens — so a thinking model's line does not collapse; the table ranks on visible tokens · one cell per model, showing its best-covered provider · models with at least 4 samples in the window first, fastest of those at the top · shared vertical scale, 0 to 71 tok/s (99th percentile) · dashed rule is the model's own mean

Every lane measuring this model

2 measured · live first · a lane is a provider, a catalogue id and a transport
ProviderLaneAvg Toks/SecMinMaxAvg TTF (ms)n
googlevia OpenRouter55.9047.5067.301180.004
googledirect42.301.5374.402740.0037

Frequently Asked Questions

How fast is Nano Banana 2 (Gemini 3.1 Flash Image)?
The latest rolling average throughput is 55.90 tokens per second with an average time to first token of 1180.00 ms across 4 recent runs.
How often are these benchmarks updated?
Benchmarks refresh automatically whenever the monitoring cron runs. The most recent run completed on Sep 1, 2026, 03:31 AM.

Related models