MedicalBenchmark
Meta: Llama 3.2 3B Instruct provider

Llama 3.2 3B Instruct

304

#304 of 319 models in the general ranking

Cumulative performance across 3 MIR exams

Net score

129.66 pts

Accuracy

37.8%

Correct / Incorrect

227 / 292

Total Cost

$0.03

Overall Performance

(vs. average)
Accuracy

37.8%

avg: 80.6%

Net score

129.66 pts

avg: 453.30 pts

Correct

227

avg: 483

Incorrect

292

avg: 90

Total Cost

$0.03

avg: $9.58

Average response time

22.4s

avg: 17.9s

Output Tokens

429K

avg: 1.3M

Reasoning Tokens

0

avg: 898K

Average confidence

83.9%

avg: 95.4%

Breakdown by Exam

MIR 2024
306
Correct
72
Incorrect
100
Accuracy
36.0%
Net score
38.66
MIR 2025
305
Correct
64
Incorrect
106
Accuracy
32.0%
Net score
28.66
MIR 2026
301
Correct
91
Incorrect
86
Accuracy
45.5%
Net score
62.33