Mistral Large Instruct 2407 — Benchmarks

Benchmark scores for Mistral Large Instruct 2407 aggregated from public leaderboards, with how it ranks among open models. See hardware requirements for what you need to run it.

Overall rank: #49 of 78 open modelscomposite 39.6/100 across 6 benchmarks in 3 categories · methodology

Instruction Following

BenchmarkScoreOpen rankAll models
LMArena Text1314.2#94 / 182#246 / 391

Knowledge

BenchmarkScoreOpen rankAll models
MMLU-Pro65.9%#45 / 144#115 / 257
MMLU80.0%#11 / 85#25 / 136

Math

BenchmarkScoreOpen rankAll models
MATH Level 544.8%#17 / 39#68 / 107
AIME 2024/20258.5%#57 / 82#222 / 270

Reasoning

BenchmarkScoreOpen rankAll models
DTBench61.2%#44 / 67#164 / 210
GPQA Diamond49.0%#54 / 94#217 / 289
SimpleBench22.5%#21 / 25#93 / 101

Scores aggregated from public benchmark sources (each linked from the benchmark pages). llmrun does not run these benchmarks.