Mistral Large — benchmark results

February 2024 Mistral Large snapshot row. Provider: Mistral. Released 2024-02-26. Access: API.

Unified ELO 1497 ± 14, rank #809 of 1776 rated models, from 95 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
AgentLeak99.3Total Leak (self-reported)100
BlueBench - Safety88.04Score (%)94.1
InfiBench58.22Score (%)92.4
TextClass Benchmark1720.36Meta-Elo (self-reported)88.8
BiGGen-Bench3.93Average Score (1-5)88.2
SEAL - Adversarial Robustness37Score85.7
LiveBench Zebra Puzzle40Score85.4
LiveBench Summarize69.23Score83.3
LiveBench Table Reformat64Score83.3
BlueBench - Bias96.97Score (%)82.4
BlueBench - Summarization18.24Score (%)82.4
CZ-EVAL - Analytical38.52Accuracy (%)81.8

Interactive version: theaggregate.ai/model?slug=mistral-large · How the rankings work · Data refreshed daily, snapshot 2026-07-22.