Mixtral_11Bx2_MoE_19B: benchmark results

Provider: Mistral. Access: Open.

Unified ELO 1502 ± 20, rank #1193 of 2928 rated models, from 12 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Open LLM Leaderboard v1 - TruthfulQA MC272MC2 (%) (0-shot)92.9
Open LLM Leaderboard v1 - HellaSwag88.47Normalized accuracy (%) (10-shot)92.1
Open LLM Leaderboard v1 - ARC Challenge71.16Normalized accuracy (%) (25-shot)88.3
Open LLM Leaderboard v1 - WinoGrande83.27Accuracy (%) (5-shot)87.9
Open LLM Leaderboard v1 - MMLU66.31Accuracy (%) (5-shot)86.7
Open LLM Leaderboard v1 - GSM8K65.28Accuracy (%) (5-shot)79.7
Open LLM Leaderboard - MuSR42.97Score70.1
Open LLM Leaderboard - BBH52.09Score59.7
Open LLM Leaderboard - MMLU-Pro33.11Score46.5
Open LLM Leaderboard - GPQA29.03Score45
Open LLM Leaderboard - IFEval38.51Score37.9
Open LLM Leaderboard - MATH Level 56.72Score36.4

Interactive version: theaggregate.ai/model?slug=mixtral-11bx2-moe-19b · How It Works · Data refreshed daily, snapshot 2026-09-23.