Go
gemma-4-31b-it
Benchmark & Insights
Google Gemini API
Updated Oct 4, 2026 All models
Sample size
232 runs
in window
Accuracy
76.8%
reference match · 250d
Confidence
100%
over 252 runs
Window end
Oct 4, 2026
most recent run
Input price
$0.13/MTok
prompt tokens
Output price
$0.38/MTok
completion tokens
Model insights
- 01 Cheap and reliable: at $0.13/$0.38 it beats 30 of 35 same-day peers and misses fewer "unsafe" days (5 of 25) than any other full-period leader.
- 02 The trade-off is a few false alarms on "MEDIUM_SAFE" days.
- 03 Its one weak spot was the late-September storm, where it called two "HIGH" days "medium".
Notes
Open-weight model; also available free via OpenRouter
Recent forecasts