MedicalBenchmark
OpenAI: GPT-5 Chat provider

GPT-5 Chat

42

#42 of 319 models in the general ranking

Cumulative performance across 3 MIR exams

Net score

565.66 pts

Accuracy

95.7%

Correct / Incorrect

574 / 25

Total Cost

$3.28

Overall Performance

(vs. average)
Accuracy

95.7%

avg: 80.6%

Net score

565.66 pts

avg: 453.30 pts

Correct

574

avg: 483

Incorrect

25

avg: 90

Total Cost

$3.28

avg: $9.58

Average response time

4.0s

avg: 17.9s

Output Tokens

292K

avg: 1.3M

Reasoning Tokens

0

avg: 898K

Average confidence

99.7%

avg: 95.4%

Breakdown by Exam

MIR 2024
9
Correct
195
Incorrect
5
Accuracy
97.5%
Net score
193.33
MIR 2025
70
Correct
184
Incorrect
15
Accuracy
92.0%
Net score
179.00
MIR 2026
35
Correct
195
Incorrect
5
Accuracy
97.5%
Net score
193.33