Devstral Small (May '25): benchmark results

Provider: Mistral. Released 2025-05-21.

Aggregate Rating 1498 ± 42, rank #2393 of 10916 rated models, from 10 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Tau2-Bench Telecom38.01Success Rate (%)47.3
AA TAU-2 Bench38.01Accuracy (%)46
AA Terminal-Bench Hard6.06Accuracy (%)33.9
AA Long Context Reasoning32Accuracy (%)29.4
AA LiveCodeBench25.82Pass@1 (%)27.2
AA MMLU-Pro63.17Accuracy (%)24.4
AA CritPt0Accuracy (%)20.9
AA GPQA Diamond43.43Accuracy (%)18.9
AA IFBench31.63Accuracy (%)17.4
AA Humanity's Last Exam4.03Accuracy (%)12.2

Interactive version: theaggregate.ai/model?slug=devstral-small-may-25 · How It Works · Data refreshed daily, snapshot 2026-11-08.