An
claude-haiku-4-5
Benchmark & Insights
Anthropic Claude API
Updated Oct 4, 2026 All models
Sample size
232 runs
in window
Accuracy
74.1%
reference match · 251d
Confidence
93%
over 252 runs
Window end
Oct 4, 2026
most recent run
Input price
$1.00/MTok
prompt tokens
Output price
$5.00/MTok
completion tokens
Model insights
- 01 Strong in calm weather: it was the best model over gemini-3.1-flash-lite-preview's spring stretch, scoring 83-84% in April and May.
- 02 Over the full period it leans low and missed 11 "unsafe" days, including both late-September "HIGH" days.
- 03 At $1/$5 it is a fair mid-tier option.
Recent forecasts