Compare AI models

Specifications
Parameters
7B/13B/33B/65B51B
Context window
2k128k
Benchmarks
ARC-Challenge
56%69.2%
HellaSwag
84.2%85.58%
WinoGrande
77%84.53%
MMLU
63.4%80.2%
GSM8K
50.9%91.43%
Overview
CompanyMetaNVIDIA
Release dateFeb 24 2023Sep 23 2024
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Llama 3.1 Nemotron 51B leads LLaMA 1 on 5 of the 5 benchmarks they both report. LLaMA 1 shipped 577 days before Llama 3.1 Nemotron 51B, so benchmark comparisons should account for the intervening progress.

LLaMA 1 has 7B/13B/33B/65B parameters, while Llama 3.1 Nemotron 51B has 51B. Context windows are 2k (LLaMA 1) vs 128k (Llama 3.1 Nemotron 51B).

On ARC-Challenge, Llama 3.1 Nemotron 51B leads at 69.2% vs LLaMA 1 at 56%. On HellaSwag, Llama 3.1 Nemotron 51B leads at 85.58% vs LLaMA 1 at 84.2%. On WinoGrande, Llama 3.1 Nemotron 51B leads at 84.53% vs LLaMA 1 at 77%. On MMLU, Llama 3.1 Nemotron 51B leads at 80.2% vs LLaMA 1 at 63.4%. On GSM8K, Llama 3.1 Nemotron 51B leads at 91.43% vs LLaMA 1 at 50.9%.