Compare AI models

Specifications
Parameters
7B/13B/70B51B
Context window
4k128k
Benchmarks
ARC-Challenge
57.4%69.2%
HellaSwag
85.3%85.58%
WinoGrande
80.2%84.53%
MMLU
68.9%80.2%
GSM8K
56.8%91.43%
Overview
CompanyMetaNVIDIA
Release dateJul 18 2023Sep 23 2024
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Llama 3.1 Nemotron 51B leads LLaMA 2 on 5 of the 5 benchmarks they both report. LLaMA 2 shipped 433 days before Llama 3.1 Nemotron 51B, so benchmark comparisons should account for the intervening progress.

LLaMA 2 has 7B/13B/70B parameters, while Llama 3.1 Nemotron 51B has 51B. Context windows are 4k (LLaMA 2) vs 128k (Llama 3.1 Nemotron 51B).

On ARC-Challenge, Llama 3.1 Nemotron 51B leads at 69.2% vs LLaMA 2 at 57.4%. On HellaSwag, Llama 3.1 Nemotron 51B leads at 85.58% vs LLaMA 2 at 85.3%. On WinoGrande, Llama 3.1 Nemotron 51B leads at 84.53% vs LLaMA 2 at 80.2%. On MMLU, Llama 3.1 Nemotron 51B leads at 80.2% vs LLaMA 2 at 68.9%. On GSM8K, Llama 3.1 Nemotron 51B leads at 91.43% vs LLaMA 2 at 56.8%.