Compare AI models

Specifications
Parameters
16B7B
Context window
32k32k
Benchmarks
HumanEval
57.3%38.4%
MATH
27.9%13.2%
Overview
CompanyDeepSeekMistral
Release dateMay 16 2024May 22 2024
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

DeepSeek V2 Lite leads Mistral 7B v0.3 on 2 of the 2 benchmarks they both report (HumanEval, MATH). DeepSeek V2 Lite shipped 6 days before Mistral 7B v0.3, so benchmark comparisons should account for the intervening progress.

DeepSeek V2 Lite has 16B parameters, while Mistral 7B v0.3 has 7B. Context windows are 32k (DeepSeek V2 Lite) vs 32k (Mistral 7B v0.3).

On HumanEval, DeepSeek V2 Lite leads at 57.3% vs Mistral 7B v0.3 at 38.4%. On MATH, DeepSeek V2 Lite leads at 27.9% vs Mistral 7B v0.3 at 13.2%.