Compare AI models

Specifications
Parameters
16B8B
Context window
32k128k
Benchmarks
HumanEval
57.3%76.8%
MATH
27.9%54.5%
Overview
CompanyDeepSeekMistral
Release dateMay 16 2024Oct 16 2024
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Ministral 8B leads DeepSeek V2 Lite on 2 of the 2 benchmarks they both report (HumanEval, MATH). DeepSeek V2 Lite shipped 153 days before Ministral 8B, so benchmark comparisons should account for the intervening progress.

DeepSeek V2 Lite has 16B parameters, while Ministral 8B has 8B. Context windows are 32k (DeepSeek V2 Lite) vs 128k (Ministral 8B).

On HumanEval, Ministral 8B leads at 76.8% vs DeepSeek V2 Lite at 57.3%. On MATH, Ministral 8B leads at 54.5% vs DeepSeek V2 Lite at 27.9%.