Compare AI models

Specifications
Parameters
236B—
Context window
128k256k
Benchmarks
HumanEval
84.76%86.6%
Overview
CompanyDeepSeekMistral
Release dateJun 28 2024Jan 13 2025
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

Codestral 25.01 leads DeepSeek V2-0628 on 1 of the 1 benchmark they both report (HumanEval). DeepSeek V2-0628 shipped 199 days before Codestral 25.01, so benchmark comparisons should account for the intervening progress.

Context windows are 128k (DeepSeek V2-0628) vs 256k (Codestral 25.01). DeepSeek V2-0628 is open weight, while Codestral 25.01 is closed.

On HumanEval, Codestral 25.01 leads at 86.6% vs DeepSeek V2-0628 at 84.76%.