MedicalBenchmark
Anthropic: Claude Sonnet 4 provider

Claude Sonnet 4

63

#63 of 319 models in the general ranking

Cumulative performance across 3 MIR exams

Net score

557.33 pts

Accuracy

94.7%

Correct / Incorrect

568 / 32

Total Cost

$6.64

Overall Performance

(vs. average)
Accuracy

94.7%

avg: 80.6%

Net score

557.33 pts

avg: 453.30 pts

Correct

568

avg: 483

Incorrect

32

avg: 90

Total Cost

$6.64

avg: $9.58

Average response time

10.4s

avg: 17.9s

Output Tokens

370K

avg: 1.3M

Reasoning Tokens

0

avg: 898K

Average confidence

99.9%

avg: 95.4%

Breakdown by Exam

MIR 2024
76
Correct
190
Incorrect
10
Accuracy
95.0%
Net score
186.66
MIR 2025
60
Correct
186
Incorrect
14
Accuracy
93.0%
Net score
181.33
MIR 2026
64
Correct
192
Incorrect
8
Accuracy
96.0%
Net score
189.33