Llama-3.1-Nemotron-70B-Instruct-HF

nvidia

Parameters

70.6B

Downloads

—

Likes

—

Avg score

36.9

6 benchmarks

Weights

Open

Released

—

About

Llama-3.1-Nemotron-70B-Instruct-HF is a 70.6B-parameter open-weights model in the Llama family from nvidia. Benchmarked on 6 evals, averaging 36.9.

llama3.1

Benchmark scores · 6

avg 36.9 — higher is better

IFEvalOpen LLM LB · 2026-05

73.8

BIG-Bench HardOpen LLM LB · 2026-05

47.1

MMLU-ProOpen LLM LB · 2026-05

43.5

MATH Lvl 5Open LLM LB · 2026-05

42.7

MuSROpen LLM LB · 2026-05

13.2

GPQA DiamondOpen LLM LB · 2026-05

1.1

Details

Publishernvidia

FamilyLlama

Parameters70.6B

Licensellama3.1

WeightsOpen

Released—

History

Capability over time

Append-only ledger — every observed change to this model's numbers.