Compare AI models

Specifications
Parameters
—1T
Context window
128k256k
API pricing
Cheapest input
—$0.60Novita
Cheapest output
—$2.50Novita
—$245Novita · bf16
Benchmarks
SWE-Bench Verified
61.6%71.3%
Overview
CompanyMistralMoonshot AI
Release dateJul 10 2025Nov 6 2025
AccessClosedOpen Weight
Model detailsView modelView model

Frequently asked questions

Kimi K2 Thinking leads Devstral Medium on 1 of the 1 benchmark they both report (SWE-Bench Verified). Devstral Medium shipped 119 days before Kimi K2 Thinking, so benchmark comparisons should account for the intervening progress.

Context windows are 128k (Devstral Medium) vs 256k (Kimi K2 Thinking). Devstral Medium is closed, while Kimi K2 Thinking is open weight.

On SWE-Bench Verified, Kimi K2 Thinking leads at 71.3% vs Devstral Medium at 61.6%.