Mistral — benchmark results

Provider: Mistral. Access: Open.

Unified ELO 895 ± 125, rank #1726 of 1726 rated models, from 14 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
MMedBench60.73Accuracy (%)75
MMedBench - Rationale (BLEU-1)45.37Accuracy (%)66.7
MedS-Bench - Text Summarization24.48Score63.6
ClinicBench37.81Average Score54.5
MedS-Bench - Concept Explanation13.53Score54.5
MedS-Bench - Text Classification48.73Score54.5
MedS-Bench - Diagnosis34.8Score45.5
MedS-Bench - Information Extraction70.18Score27.3
MedS-Bench - Clinical Outcome Prediction50.14Score22.7
BIRD-SQL55.84Execution Accuracy (%)22.2
MedS-Bench - Treatment Planning38.93Score18.2
RULER68.4Avg Accuracy (%)15

Interactive version: aibenchmarks.dev/model?slug=mistral · How the rankings work · Data refreshed daily, snapshot 2026-07-08.