Skip to content
Trending storyDeveloping

Mistral Large 4 Claims Strong STEMBench Results in Engineering and Astronomy

1 article1 sourcesince Oct 16Last article 59m ago ·

Overview

AISummary of 1 article

Mistral AI says Mistral Large 4 performs well in engineering and astronomy questions on STEMBench, a private benchmark run by Ulam.

The company's X post, dated October 16, 2026, links to Ulam's STEMBench page as evidence but gives no scores or comparisons with other models.

Written by AI from the articles below · updated Oct 16, 7:37 AM ET

Check the sources:

Article timeline

The articles in this story. Times are ET.

Oct 16
  1. Mistral AIOfficial
    Mistral Large 4 scores well on STEMBench engineering and astronomy questions

    AIMistral AI says Mistral Large 4 performs well in engineering and astronomy on Ulam's private benchmark, STEMBench. The post gives no scores or comparisons.

    Image from @MistralAI's post

Heat trend

Not enough continuous observations to show a trend yet.