MedicalBenchmark
OpenAI: GPT-4o provider

GPT-4o

126

#126 of 319 models in the general ranking

Cumulative performance across 3 MIR exams

Net score

526.33 pts

Accuracy

89.7%

Correct / Incorrect

538 / 35

Total Cost

$3.05

Overall Performance

(vs. average)
Accuracy

89.7%

avg: 80.6%

Net score

526.33 pts

avg: 453.30 pts

Correct

538

avg: 483

Incorrect

35

avg: 90

Total Cost

$3.05

avg: $9.58

Average response time

7.5s

avg: 17.9s

Output Tokens

231K

avg: 1.3M

Reasoning Tokens

0

avg: 898K

Average confidence

97.8%

avg: 95.4%

Breakdown by Exam

MIR 2024
113
Correct
184
Incorrect
7
Accuracy
92.0%
Net score
181.66
MIR 2025
150
Correct
171
Incorrect
21
Accuracy
85.5%
Net score
164.00
MIR 2026
114
Correct
183
Incorrect
7
Accuracy
91.5%
Net score
180.66