Llama 4 Behemoth: benchmark results
Provider: Meta.
Aggregate Rating 1658 ± 34, rank #572 of 10916 rated models, from 5 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MMMU Benchmark | 76.1 | Validation Score | 89.4 |
| Vellum - MATH | 95 | Accuracy (%) | 65 |
| Vellum - MMMLU | 85.8 | Accuracy (%) | 45.5 |
| Vellum - GPQA | 73.7 | Accuracy (%) | 35.7 |
| Vellum - LiveCodeBench | 49.4 | Pass@1 (%) | 18.2 |
Interactive version: theaggregate.ai/model?slug=llama-4-behemoth · How It Works · Data refreshed daily, snapshot 2026-11-08.