MedicalBenchmark
NVIDIA: Llama 3.1 Nemotron 70B Instruct provider

Llama 3.1 Nemotron 70B Instruct

194

#194 of 319 models in the general ranking

Cumulative performance across 3 MIR exams

Net score

478.66 pts

Accuracy

84.2%

Correct / Incorrect

505 / 79

Total Cost

$0.87

Overall Performance

(vs. average)
Accuracy

84.2%

avg: 80.6%

Net score

478.66 pts

avg: 453.30 pts

Correct

505

avg: 483

Incorrect

79

avg: 90

Total Cost

$0.87

avg: $9.58

Average response time

14.4s

avg: 17.9s

Output Tokens

432K

avg: 1.3M

Reasoning Tokens

0

avg: 898K

Average confidence

96.1%

avg: 95.4%

Breakdown by Exam

MIR 2024
214
Correct
165
Incorrect
30
Accuracy
82.5%
Net score
155.00
MIR 2025
206
Correct
160
Incorrect
33
Accuracy
80.0%
Net score
149.00
MIR 2026
145
Correct
180
Incorrect
16
Accuracy
90.0%
Net score
174.66