Compare AI models

Specifications
Context window
200k128k
Benchmarks
SWE-Bench Verified
7.2%61.6%
Overview
CompanyAnthropicMistral
Release dateMar 4 2024Jul 10 2025
AccessClosedClosed
Model detailsView modelView model

Frequently asked questions

Devstral Medium leads Claude 3 Haiku on 1 of the 1 benchmark they both report (SWE-Bench Verified). Claude 3 Haiku shipped 493 days before Devstral Medium, so benchmark comparisons should account for the intervening progress.

Context windows are 200k (Claude 3 Haiku) vs 128k (Devstral Medium).

On SWE-Bench Verified, Devstral Medium leads at 61.6% vs Claude 3 Haiku at 7.2%.