Go
gemini-3.6-flash
Benchmark & Insights
Google Gemini API
Updated Sep 12, 2026 All models
Sample size
211 runs
in window
Accuracy
86.4%
consensus match · 214d
Confidence
90%
over 214 runs
Window end
Sep 12, 2026
most recent run
Input price
$1.50/MTok
prompt tokens
Output price
$7.50/MTok
completion tokens
Model insights
- 01 The newest Flash is a step back from gemini-3.5-flash despite similar pricing.
- 02 It leans hard optimistic, calling "low" on "medium" days more than any Gemini except the Pro preview, and it said "safe" on four "unsafe" days.
- 03 Misses cluster in March and August.
Notes
20% cheaper than the previous version in output tokens. Gemini 3.6 Flash Model Card