vicuna-7B-physics: benchmark results

Provider: LMSYS. Access: Open.

Aggregate Rating 1284 ± 15, rank #8937 of 10916 rated models, from 68 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
MMLU-by-task - High School Mathematics31.11Accuracy (%)83.3
MMLU-by-task - Medical Genetics60Accuracy (%)78.5
MMLU-by-task - TruthfulQA MC249.31Accuracy (%)72.8
MMLU-by-task - College Physics29.41Accuracy (%)71.1
MMLU-by-task - TruthfulQA MC131.82Accuracy (%)65.9
MMLU-by-task - Security Studies58.78Accuracy (%)63.9
MMLU-by-task - US Foreign Policy76Accuracy (%)63.7
MMLU-by-task - Business Ethics52Accuracy (%)61.8
MMLU-by-task - College Mathematics33Accuracy (%)61.7
MMLU-by-task - Clinical Knowledge53.96Accuracy (%)61.2
MMLU-by-task - Jurisprudence59.26Accuracy (%)60.7
MMLU-by-task - Virology42.77Accuracy (%)60.3

Interactive version: theaggregate.ai/model?slug=vicuna-7b-physics · How It Works · Data refreshed daily, snapshot 2026-11-08.