MedicalBenchmark
OpenAI: GPT-3.5 Turbo provider

GPT-3.5 Turbo

270

#270 of 319 models in the general ranking

Cumulative performance across 3 MIR exams

Net score

299.00 pts

Accuracy

62.2%

Correct / Incorrect

373 / 222

Total Cost

$0.42

Overall Performance

(vs. average)
Accuracy

62.2%

avg: 80.6%

Net score

299.00 pts

avg: 453.30 pts

Correct

373

avg: 483

Incorrect

222

avg: 90

Total Cost

$0.42

avg: $9.58

Average response time

2.7s

avg: 17.9s

Output Tokens

185K

avg: 1.3M

Reasoning Tokens

0

avg: 898K

Average confidence

98.6%

avg: 95.4%

Breakdown by Exam

MIR 2024
268
Correct
129
Incorrect
70
Accuracy
64.5%
Net score
105.66
MIR 2025
274
Correct
112
Incorrect
87
Accuracy
56.0%
Net score
83.00
MIR 2026
272
Correct
132
Incorrect
65
Accuracy
66.0%
Net score
110.33