MedicalBenchmark
Sao10K: Llama 3 8B Lunaris provider

Llama 3 8B Lunaris

288

#288 of 319 models in the general ranking

Cumulative performance across 3 MIR exams

Net score

198.00 pts

Accuracy

47.7%

Correct / Incorrect

286 / 264

Total Cost

$0.02

Overall Performance

(vs. average)
Accuracy

47.7%

avg: 80.6%

Net score

198.00 pts

avg: 453.30 pts

Correct

286

avg: 483

Incorrect

264

avg: 90

Total Cost

$0.02

avg: $9.58

Average response time

4.8s

avg: 17.9s

Output Tokens

213K

avg: 1.3M

Reasoning Tokens

0

avg: 898K

Average confidence

89.9%

avg: 95.4%

Breakdown by Exam

MIR 2024
297
Correct
88
Incorrect
104
Accuracy
44.0%
Net score
53.33
MIR 2025
289
Correct
86
Incorrect
92
Accuracy
43.0%
Net score
55.33
MIR 2026
285
Correct
112
Incorrect
68
Accuracy
56.0%
Net score
89.33