MedicalBenchmark
OpenAI: GPT-4 Turbo Preview provider

GPT-4 Turbo Preview

148

#148 of 319 models in the general ranking

Cumulative performance across 3 MIR exams

Net score

511.00 pts

Accuracy

88.5%

Correct / Incorrect

531 / 60

Total Cost

$13.10

Overall Performance

(vs. average)
Accuracy

88.5%

avg: 80.6%

Net score

511.00 pts

avg: 453.30 pts

Correct

531

avg: 483

Incorrect

60

avg: 90

Total Cost

$13.10

avg: $9.58

Average response time

13.0s

avg: 17.9s

Output Tokens

339K

avg: 1.3M

Reasoning Tokens

0

avg: 898K

Average confidence

97.1%

avg: 95.4%

Breakdown by Exam

MIR 2024
147
Correct
181
Incorrect
17
Accuracy
90.5%
Net score
175.33
MIR 2025
147
Correct
173
Incorrect
25
Accuracy
86.5%
Net score
164.66
MIR 2026
159
Correct
177
Incorrect
18
Accuracy
88.5%
Net score
171.00