MedicalBenchmark
OpenAI: GPT-4 provider

GPT-4

155

#155 of 319 models in the general ranking

Cumulative performance across 3 MIR exams

Net score

507.00 pts

Accuracy

87.7%

Correct / Incorrect

526 / 57

Total Cost

$25.07

Overall Performance

(vs. average)
Accuracy

87.7%

avg: 80.6%

Net score

507.00 pts

avg: 453.30 pts

Correct

526

avg: 483

Incorrect

57

avg: 90

Total Cost

$25.07

avg: $9.58

Average response time

11.1s

avg: 17.9s

Output Tokens

272K

avg: 1.3M

Reasoning Tokens

0

avg: 898K

Average confidence

96.4%

avg: 95.4%

Breakdown by Exam

MIR 2024
157
Correct
178
Incorrect
15
Accuracy
89.0%
Net score
173.00
MIR 2025
127
Correct
175
Incorrect
21
Accuracy
87.5%
Net score
168.00
MIR 2026
179
Correct
173
Incorrect
21
Accuracy
86.5%
Net score
166.00