Compare AI models

Specifications
Parameters
236B8B
Context window
128k4k
Benchmarks
MMLU
77.8%54.4%
Overview
CompanyDeepSeekNVIDIA
Release dateMay 6 2024Nov 15 2023
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

DeepSeek V2 leads Nemotron-3 8B on 1 of the 1 benchmark they both report (MMLU). Nemotron-3 8B shipped 173 days before DeepSeek V2, so benchmark comparisons should account for the intervening progress.

DeepSeek V2 has 236B parameters, while Nemotron-3 8B has 8B. Context windows are 128k (DeepSeek V2) vs 4k (Nemotron-3 8B).

On MMLU, DeepSeek V2 leads at 77.8% vs Nemotron-3 8B at 54.4%.