Compare AI models

Specifications
Parameters
—6B
Context window
128k32k
API pricing
Cheapest input
$0.40Mistral—
Cheapest output
$2.00Mistral—
$180Mistral · zdr—
Benchmarks
HumanEval
92.1%9.8%
Overview
CompanyMistralZ.ai
Release dateMay 7 2025Jun 25 2023
AccessClosedOpen Weight
Model detailsView modelView model

Frequently asked questions

Mistral Medium 3 leads ChatGLM2-6B on 1 of the 1 benchmark they both report (HumanEval). ChatGLM2-6B shipped 682 days before Mistral Medium 3, so benchmark comparisons should account for the intervening progress.

Context windows are 128k (Mistral Medium 3) vs 32k (ChatGLM2-6B). Mistral Medium 3 is closed, while ChatGLM2-6B is open weight.

On HumanEval, Mistral Medium 3 leads at 92.1% vs ChatGLM2-6B at 9.8%.