Vicuna-7B-v1.5-16k: benchmark results
Provider: LMSYS. Released 2023-08-01. Access: Open.
Unified ELO 1298 ± 12, rank #1410 of 1605 rated models, from 140 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MMLU-by-task - Econometrics | 39.47 | Accuracy (%) | 91.5 |
| MMLU-by-task - Machine Learning | 39.29 | Accuracy (%) | 79 |
| MMLU-by-task - TruthfulQA MC2 | 50.35 | Accuracy (%) | 77.7 |
| MMLU-by-task - Business Ethics | 55 | Accuracy (%) | 74.5 |
| MMLU-by-task - TruthfulQA MC1 | 33.41 | Accuracy (%) | 72.8 |
| MMLU-by-task - Medical Genetics | 58 | Accuracy (%) | 71.8 |
| MMLU-by-task - Virology | 45.18 | Accuracy (%) | 70.1 |
| MMLU-by-task - Professional Accounting | 39.72 | Accuracy (%) | 68.3 |
| MMLU-by-task - Conceptual Physics | 42.98 | Accuracy (%) | 68 |
| MMLU-by-task - Human Aging | 60.99 | Accuracy (%) | 65.4 |
| MMLU-by-task - Security Studies | 59.18 | Accuracy (%) | 64.6 |
| MMLU-by-task - Nutrition | 56.21 | Accuracy (%) | 64.1 |
Interactive version: theaggregate.ai/model?slug=vicuna-7b-v1-5-16k · How It Works · Data refreshed daily, snapshot 2026-09-26.