Mistral Small (Sep '24) — benchmark results

Provider: Mistral. Released 2024-09-17. Access: Open.

Unified ELO 1363 ± 45, rank #1424 of 1841 rated models, from 7 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
AA MATH-50056.27Accuracy (%)19.4
AA Humanity's Last Exam4.32Accuracy (%)18.9
AA MMLU-Pro52.93Accuracy (%)16.6
AA GPQA Diamond38.08Accuracy (%)15
AA LiveCodeBench14.07Pass@1 (%)14.3
Epoch AI - Scicode15.62Score14.2
Artificial Analysis Intelligence Index4.66Intelligence Index13.7

Interactive version: theaggregate.ai/model?slug=mistral-small-sep-24 · How It Works · Data refreshed daily, snapshot 2026-07-25.