Llama 2 22B Daydreamer V3: benchmark results
Provider: Meta. Access: Open.
Aggregate Rating 1318 ± 15, rank #8361 of 10916 rated models, from 68 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MMLU-by-task - High School Physics | 38.41 | Accuracy (%) | 90.5 |
| MMLU-by-task - College Chemistry | 46 | Accuracy (%) | 90.3 |
| MMLU-by-task - College Physics | 34.31 | Accuracy (%) | 86.4 |
| MMLU-by-task - High School Mathematics | 31.11 | Accuracy (%) | 83.3 |
| MMLU-by-task - Medical Genetics | 62 | Accuracy (%) | 82.4 |
| MMLU-by-task - Abstract Algebra | 34 | Accuracy (%) | 82 |
| MMLU-by-task - College Computer Science | 47 | Accuracy (%) | 80.1 |
| MMLU-by-task - High School Statistics | 45.83 | Accuracy (%) | 76.9 |
| MMLU-by-task - Formal Logic | 36.51 | Accuracy (%) | 76.5 |
| MMLU-by-task - Global Facts | 37 | Accuracy (%) | 76.3 |
| MMLU-by-task - Anatomy | 49.63 | Accuracy (%) | 75.9 |
| MMLU-by-task - High School European History | 68.48 | Accuracy (%) | 75.5 |
Interactive version: theaggregate.ai/model?slug=llama-2-22b-daydreamer-v3 · How It Works · Data refreshed daily, snapshot 2026-11-08.