0

Llama 3.1 Nemotron Instruct 70B

Closed

Llama 3.1 Nemotron Instruct 70B is an AI model from NVIDIA.

Publisher
NVIDIA
Family
Llama
Context
131K
Input
Output
License
Llama 3.1
Released
15 Oct 2024

Cite

Notes

Only stored in your browser.

Attribution

Benchmark scores
AALiveBench
Attribution policy →
Context
131K
Input
$1.20
/1M tok
Output
$1.20
/1M tok
Blended
$1.20
#319/ 414/1M · 3:1
Speed
294
#10/ 414tok/s

Price & speed via Artificial Analysis · ranked across tracked models

Reported on 15 evals across 7 domains - best 73.3% on MATH-500 · #109 of 178

Reported eval scores

15
LiveBench - Instruction Following69.7
Terminal-Bench (Hard)4.5
LiveBench - Language34.2
LiveBench - Coding32.8
IFBench30.7
AIME 2024: Problems from the American Invitational Mathematics Examination24.7
MATH-50073.3
MMLU-Pro69
Humanity's Last Exam (HLE)4.6
LiveBench45.5
GPQA Diamond46.5
SciCode23.3
τ²-bench (Tau²-bench)23.1
LiveCodeBench16.9
AIME 2025: Problems from the American Invitational Mathematics Examination11