MedicalBenchmark
DeepSeek: R1 Distill Llama 70B provider

R1 Distill Llama 70B

187

#187 of 319 models in the general ranking

Cumulative performance across 3 MIR exams

Net score

486.00 pts

Accuracy

85.5%

Correct / Incorrect

513 / 81

Total Cost

$0.81

Overall Performance

(vs. average)
Accuracy

85.5%

avg: 80.6%

Net score

486.00 pts

avg: 453.30 pts

Correct

513

avg: 483

Incorrect

81

avg: 90

Total Cost

$0.81

avg: $9.58

Average response time

14.6s

avg: 17.9s

Output Tokens

531K

avg: 1.3M

Reasoning Tokens

449K

avg: 898K

Average confidence

98.7%

avg: 95.4%

Breakdown by Exam

MIR 2024
185
Correct
175
Incorrect
24
Accuracy
87.5%
Net score
167.00
MIR 2025
184
Correct
166
Incorrect
31
Accuracy
83.0%
Net score
155.66
MIR 2026
197
Correct
172
Incorrect
26
Accuracy
86.0%
Net score
163.33