NVIDIA/Dense

Llama-3.1-Nemotron-70B

Name: Llama-3.1-Nemotron-70B
Author: NVIDIA

NVIDIA's Llama 3.1 70B fine-tune. Top-tier instruction following and helpfulness.

chatreasoning

70.6B

Parameters

128K

Context length

Benchmarks

Quantizations

200K

HF downloads

Architecture

Dense

Released

2024-10-03

Layers

KV Heads

Head Dim

128

Family

nemotron

Quantization Options

Quant	Bits	VRAM	Quality
IQ2_M	2.93	26.3 GB	low
Q2_K	3.16	28.4 GB	low
IQ3_XXS	3.25	29.2 GB	low
IQ3_XS	3.5	31.4 GB	low
Q3_K_S	3.64	32.6 GB	low
IQ3_M	3.76	33.7 GB	low
Q3_K_M	4	35.8 GB	low
Q3_K_L	4.3	38.4 GB	moderate
IQ4_XS	4.46	39.8 GB	moderate
Q4_K_S	4.67	41.7 GB	moderate
Q4_K_M	4.89	43.6 GB	good
Q5_K_S	5.57	49.6 GB	good
Q5_K_M	5.7	50.8 GB	good
Q6_K	6.56	58.4 GB	excellent
Q8_0	8.5	75.5 GB	lossless
FP16	16	141.7 GB	lossless

Select your GPU above to see speed estimates and compatibility for each quantization.

▸ READY TO RUN THIS?RENT BY THE HOUR

RENT A GPU AND RUN LLAMA-3.1-NEMOTRON-70B NOW

Rent on RunPod →Or Vast.ai →

Spin up an A100 / H100 / 4090 in ~60s. Pay by the second. Cancel anytime.

Community Ratings

Loading ratings...

Benchmarks (14)

Arena Elo1284

IFEval85.0

BBH80.0

HumanEval77.0

MATH72.0

MMLU-PRO60.0

GPQA51.9

GPQA Diamond46.5

BigCodeBench38.7

LiveCodeBench16.9

MUSR13.2

AIME11.0

MATH-50011.0

HLE4.6

Run this model

▸Easiest way to get started·Beginners

DOCS ↗

curl -fsSL https://ollama.com/install.sh | sh

$ollama run nemotron:71b-q4_K_M

Tag may need adjustment — check ollama.com/library/nemotron for available tags.

▸ SETUP GUIDE

Auto-setup with fitmyllm CLI

Detects your GPU, recommends the best model, downloads it, and starts chatting — zero config. Benchmarks your speed and contributes anonymous data to improve predictions.

pip install fitmyllmthen run fitmyllmLearn more

Auto-detect GPULive tok/s in chatSpeed benchmarks9 inference engines

HuggingFace GGUF Downloads Build Hardware