An
claude-opus-4-8
Benchmark & Insights
Anthropic Claude API
Updated Oct 4, 2026 All models
Sample size
232 runs
in window
Accuracy
76.5%
reference match · 251d
Confidence
83%
over 252 runs
Window end
Oct 4, 2026
most recent run
Input price
$5.00/MTok
prompt tokens
Output price
$25.00/MTok
completion tokens
Model insights
- 01 Ranks well on overall match but misses 14 of 25 "unsafe" days, more than most models above and near it.
- 02 Never raises a false alarm and leans low, calling 31 "MEDIUM_SAFE" days "low" and both late-September "HIGH" days "medium".
- 03 At $5/$25 it trails claude-sonnet-5-5, which costs less than half as much.
Notes
Improved version of Opus 4.7. Release notes for Claude Opus 4.8
Recent forecasts