Compare AI models

Specifications
Parameters
16B51B
Context window
4k128k
Benchmarks
MMLU
47.2%80.2%
GSM8K
62.2%91.43%
Overview
CompanyDeepSeekNVIDIA
Release dateJan 11 2024Sep 23 2024
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Llama 3.1 Nemotron 51B leads DeepSeek-MoE on 2 of the 2 benchmarks they both report (MMLU, GSM8K). DeepSeek-MoE shipped 256 days before Llama 3.1 Nemotron 51B, so benchmark comparisons should account for the intervening progress.

DeepSeek-MoE has 16B parameters, while Llama 3.1 Nemotron 51B has 51B. Context windows are 4k (DeepSeek-MoE) vs 128k (Llama 3.1 Nemotron 51B).

On MMLU, Llama 3.1 Nemotron 51B leads at 80.2% vs DeepSeek-MoE at 47.2%. On GSM8K, Llama 3.1 Nemotron 51B leads at 91.43% vs DeepSeek-MoE at 62.2%.