An
claude-sonnet-5-5
Benchmark & Insights
Anthropic Claude API
Updated Oct 4, 2026 All models
Sample size
232 runs
in window
Accuracy
81.2%
reference match · 282d
Confidence
84%
over 282 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 The clear leader: on the full period it beats every one of its 35 same-day peers.
- 02 It never raised a false alarm and gets nearly every "LOW" day and all 6 "HIGH" days right; its 10 misses are all hard "MEDIUM_UNSAFE" days most models also called "safe".
- 03 At $2/$10 it beats every Opus model at half their price or less.
Recent forecasts