WizardLM 2 8x22B — Benchmarks

Benchmark scores for WizardLM 2 8x22B aggregated from public leaderboards, with how it ranks among open models. See hardware requirements for what you need to run it.

Overall rank: #64 of 73 open modelscomposite 27.6/100 across 3 benchmarks in 3 categories · methodology

Knowledge

BenchmarkScoreOpen rankAll models
MMLU-Pro39.2%#83 / 119#201 / 259

Math

BenchmarkScoreOpen rankAll models
MATH Level 525.7%#21 / 32#80 / 108

Reasoning

BenchmarkScoreOpen rankAll models
GPQA Diamond43.4%#29 / 46#140 / 182

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