Compare AI models

Specifications
Parameters
236B7B
Context window
128k32k
Benchmarks
HumanEval
84.76%38.4%
MATH
71.02%13.2%
Overview
CompanyDeepSeekMistral
Release dateJun 28 2024May 22 2024
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

DeepSeek V2-0628 leads Mistral 7B v0.3 on 2 of the 2 benchmarks they both report (HumanEval, MATH). Mistral 7B v0.3 shipped 37 days before DeepSeek V2-0628, so benchmark comparisons should account for the intervening progress.

DeepSeek V2-0628 has 236B parameters, while Mistral 7B v0.3 has 7B. Context windows are 128k (DeepSeek V2-0628) vs 32k (Mistral 7B v0.3).

On HumanEval, DeepSeek V2-0628 leads at 84.76% vs Mistral 7B v0.3 at 38.4%. On MATH, DeepSeek V2-0628 leads at 71.02% vs Mistral 7B v0.3 at 13.2%.