MedicalBenchmark
OpenAI: GPT-4o Search Preview provider

GPT-4o Search Preview

124

#124 of 319 models in the general ranking

Cumulative performance across 3 MIR exams

Net score

527.00 pts

Accuracy

90.3%

Correct / Incorrect

542 / 45

Total Cost

$0.16

Overall Performance

(vs. average)
Accuracy

90.3%

avg: 80.6%

Net score

527.00 pts

avg: 453.30 pts

Correct

542

avg: 483

Incorrect

45

avg: 90

Total Cost

$0.16

avg: $9.58

Average response time

10.5s

avg: 17.9s

Output Tokens

281K

avg: 1.3M

Reasoning Tokens

0

avg: 898K

Average confidence

97.7%

avg: 95.4%

Breakdown by Exam

MIR 2024
69
Correct
190
Incorrect
9
Accuracy
95.0%
Net score
187.00
MIR 2025
162
Correct
169
Incorrect
23
Accuracy
84.5%
Net score
161.33
MIR 2026
125
Correct
183
Incorrect
13
Accuracy
91.5%
Net score
178.66