Compare AI models

Specifications
Parameters
2B/7B51B
Context window
8k128k
Benchmarks
ARC-Challenge
53.2%69.2%
HellaSwag
81.2%85.58%
MMLU
64.3%80.2%
GSM8K
46.4%91.43%
Overview
CompanyGoogleNVIDIA
Release dateFeb 21 2024Sep 23 2024
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Llama 3.1 Nemotron 51B leads Gemma 1 on 4 of the 4 benchmarks they both report (ARC-Challenge, HellaSwag, MMLU, GSM8K). Gemma 1 shipped 215 days before Llama 3.1 Nemotron 51B, so benchmark comparisons should account for the intervening progress.

Gemma 1 has 2B/7B parameters, while Llama 3.1 Nemotron 51B has 51B. Context windows are 8k (Gemma 1) vs 128k (Llama 3.1 Nemotron 51B).

On ARC-Challenge, Llama 3.1 Nemotron 51B leads at 69.2% vs Gemma 1 at 53.2%. On HellaSwag, Llama 3.1 Nemotron 51B leads at 85.58% vs Gemma 1 at 81.2%. On MMLU, Llama 3.1 Nemotron 51B leads at 80.2% vs Gemma 1 at 64.3%. On GSM8K, Llama 3.1 Nemotron 51B leads at 91.43% vs Gemma 1 at 46.4%.