Compare AI models

Specifications
Parameters
236B—
Context window
128k128k
API pricing
Cheapest input
—$0.40Mistral
Cheapest output
—$2.00Mistral
—$180Mistral · zdr
Benchmarks
HumanEval
84.76%92.1%
Overview
CompanyDeepSeekMistral
Release dateJun 28 2024May 7 2025
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

Mistral Medium 3 leads DeepSeek V2-0628 on 1 of the 1 benchmark they both report (HumanEval). DeepSeek V2-0628 shipped 313 days before Mistral Medium 3, so benchmark comparisons should account for the intervening progress.

Context windows are 128k (DeepSeek V2-0628) vs 128k (Mistral Medium 3). DeepSeek V2-0628 is open weight, while Mistral Medium 3 is closed.

On HumanEval, Mistral Medium 3 leads at 92.1% vs DeepSeek V2-0628 at 84.76%.