Evaluates whether models can precisely count occurrences of the lowercase letter “r” in words and word-like strings while returning only the integer count.
Scorer: ExactMatch
loading chart
| # | score | model | cost price | ttft | total tok |
|---|---|---|---|---|---|
| 1 | 90% | gpt-5.2 | $0.00012 $1.75 / $14.00 per 1M | 1.21s | 38.9 |
| 2 | 70% | anthropic/claude-haiku-4.5 | $0.00006 $1.00 / $5.00 per 1M | 0.87s | 43.7 |
| 3 | 70% | anthropic/claude-opus-4.5 | $0.00032 $5.00 / $25.00 per 1M | 3.22s | 43.7 |