An
claude-sonnet-5
Benchmark & Insights
Anthropic Claude API
Updated Oct 4, 2026 All models
Sample size
232 runs
in window
Accuracy
72.9%
reference match · 251d
Confidence
85%
over 253 runs
Window end
Oct 4, 2026
most recent run
Input price
$2.00/MTok
prompt tokens
Output price
$10.00/MTok
completion tokens
Model insights
- 01 Optimistic: it rated risk too low on 47 days, called 40 of 76 "MEDIUM_SAFE" days "low", and called 6 "MEDIUM_UNSAFE" days "low", more than almost any model.
- 02 It missed 12 "unsafe" days with zero false alarms. claude-sonnet-5-5 fixes this at the same $2/$10.
Notes
Introductory pricing of $2/$10 per million input/output tokens is in effect through August 31, 2026, after which the standard pricing of $3/$15 per million input/output tokens will take effect.
Recent forecasts