MedicalBenchmark
AllenAI: Olmo 3 32B Think provider

Olmo 3 32B Think

249

#249 of 319 models in the general ranking

Cumulative performance across 3 MIR exams

Net score

399.66 pts

Accuracy

72.8%

Correct / Incorrect

437 / 112

Total Cost

$1.12

Overall Performance

(vs. average)
Accuracy

72.8%

avg: 80.6%

Net score

399.66 pts

avg: 453.30 pts

Correct

437

avg: 483

Incorrect

112

avg: 90

Total Cost

$1.12

avg: $9.58

Average response time

46.7s

avg: 17.9s

Output Tokens

2.1M

avg: 1.3M

Reasoning Tokens

1.9M

avg: 898K

Average confidence

88.0%

avg: 95.4%

Breakdown by Exam

MIR 2024
244
Correct
152
Incorrect
36
Accuracy
76.0%
Net score
140.00
MIR 2025
247
Correct
140
Incorrect
42
Accuracy
70.0%
Net score
126.00
MIR 2026
253
Correct
145
Incorrect
34
Accuracy
72.5%
Net score
133.66