Mistral Small (Feb '24) — benchmark results

Provider: Mistral. Released 2024-02-26. Access: Open.

Unified ELO 1343 ± 46, rank #1491 of 1841 rated models, from 7 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
AA Humanity's Last Exam4.4Accuracy (%)20.3
AA MATH-50056.2Accuracy (%)18.9
Epoch AI - Scicode13.43Score12.8
Artificial Analysis Intelligence Index3.62Intelligence Index9.5
AA LiveCodeBench11.11Pass@1 (%)9.1
AA MMLU-Pro41.9Accuracy (%)9
AA GPQA Diamond30.2Accuracy (%)7.6

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