Op
gpt-5-mini Benchmark & Insights
OpenAI OpenAI API
Updated Jul 18, 2026 All models
Sample size
157 runs
in window
Accuracy
87.4%
consensus match · 199d
Confidence
82%
over 205 runs
Window end
Jul 18, 2026
most recent run
Input price
$0.25/MTok
prompt tokens
Output price
$2.00/MTok
completion tokens
Model insights
- 01 A chronic worrier that bumps "low" days to "medium" (20 overratings vs 5 under), with the pattern accelerating from March onward.
- 02 At $0.25/$2 it is cheap, but gemini-3.1-flash-lite costs the same and agrees with consensus 9 points more often.
Recent forecasts
Date
Conf.
Risk
Safe
Jul 23, 2026
90%
low
Safe
Jul 22, 2026
85%
low
Safe
Jul 21, 2026
85%
high
Unsafe
Jul 20, 2026
90%
low
Safe
Jul 19, 2026
90%
medium
Safe
Jul 18, 2026
70%
high
Unsafe
Jul 17, 2026
80%
medium
Safe
Jul 16, 2026
80%
medium
Safe
Jul 15, 2026
90%
medium
Safe
Jul 14, 2026
90%
medium
Safe