LightGPT: benchmark results
Provider: Other.
Aggregate Rating 1159 ± 17, rank #9838 of 10916 rated models, from 68 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MMLU-by-task - College Mathematics | 33 | Accuracy (%) | 61.7 |
| MMLU-by-task - College Physics | 26.47 | Accuracy (%) | 57.9 |
| MMLU-by-task - Abstract Algebra | 29 | Accuracy (%) | 45.4 |
| MMLU-by-task - Security Studies | 41.22 | Accuracy (%) | 42 |
| MMLU-by-task - Professional Law | 29.53 | Accuracy (%) | 36.8 |
| MMLU-by-task - Conceptual Physics | 33.62 | Accuracy (%) | 35.8 |
| MMLU-by-task - Prehistory | 34.88 | Accuracy (%) | 35.4 |
| MMLU-by-task - High School Government And Politics | 38.34 | Accuracy (%) | 35.2 |
| MMLU-by-task - Moral Scenarios | 24.47 | Accuracy (%) | 34.7 |
| MMLU-by-task - Human Aging | 38.57 | Accuracy (%) | 34.5 |
| MMLU-by-task - Professional Accounting | 28.37 | Accuracy (%) | 34.2 |
| MMLU-by-task - Computer Security | 36 | Accuracy (%) | 33 |
Interactive version: theaggregate.ai/model?slug=lightgpt · How It Works · Data refreshed daily, snapshot 2026-11-08.