An
claude-sonnet-4-6 Benchmark & Insights
Anthropic Claude API
Updated Jul 18, 2026 All models
Sample size
157 runs
in window
Accuracy
80.0%
consensus match · 160d
Confidence
87%
over 162 runs
Window end
Jul 18, 2026
most recent run
Input price
$3.00/MTok
prompt tokens
Output price
$15.00/MTok
completion tokens
Model insights
- 01 The mirror image of claude-sonnet-5: heavily "pessimistic" (35 over vs 1 under) where its successor is purely "optimistic".
- 02 It spent Feb-Apr almost ritually calling "medium" on "low" days and raised four false "unsafe" alarms; the sonnet-5 upgrade was worth 7.7 points.
Recent forecasts