mt5-xl: benchmark results

Provider: Google. Released 2020-10-22. Access: API.

Aggregate Rating 1150 ± 50, rank #9857 of 10552 rated models, from 33 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Open LLM Leaderboard - MuSR37.95Score29.4
Open LLM Leaderboard - GPQA26.43Score18.3
Open PL LLM - KLEJ NER (multiple choice, 5-shot)28.67Accuracy (%)18.1
Open PL LLM - PolQA Reranking (multiple choice, 5-shot)53.36Accuracy (%)16
Open PL LLM - EQ-Bench0.05EQ-Bench Score13.7
Open LLM Leaderboard - IFEval19.6Score11.3
Open PL LLM - PPC (multiple choice, 5-shot)39.6Accuracy (%)10.7
Open LLM Leaderboard - BBH30.47Score6.4
Open PL LLM - PPC (generative, 5-shot)24Accuracy (%)6.4
Open PL LLM - CBD (generative, 5-shot)14.78Macro F1 (%)6.1
Open PL LLM - RAG0.08Average RAG Score (%)5.6
Open PL LLM - Generative-28.91Average Generative Score (%)4.9

Interactive version: theaggregate.ai/model?slug=mt5-xl · How It Works · Data refreshed daily, snapshot 2026-11-16.