Compare AI models

Specifications
Parameters
236B22B
Context window
128k32k
Benchmarks
HumanEval
84.76%81.1%
Overview
CompanyDeepSeekMistral
Release dateJun 28 2024May 29 2024
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

DeepSeek V2-0628 leads Codestral 22B on 1 of the 1 benchmark they both report (HumanEval). Codestral 22B shipped 30 days before DeepSeek V2-0628, so benchmark comparisons should account for the intervening progress.

DeepSeek V2-0628 has 236B parameters, while Codestral 22B has 22B. Context windows are 128k (DeepSeek V2-0628) vs 32k (Codestral 22B).

On HumanEval, DeepSeek V2-0628 leads at 84.76% vs Codestral 22B at 81.1%.