UltraLM-13B-fp16: benchmark results
TheBloke's Hugging Face publication of Open BMB's UltraLM 13B, whose original release is openbmb/UltraLM-13b. Conversion publication is credited separately from the original model creator. Provider: TheBloke. Access: Open.
Aggregate Rating 1323 ± 15, rank #8290 of 10916 rated models, from 68 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MMLU-by-task - TruthfulQA MC1 | 35.86 | Accuracy (%) | 84.5 |
| MMLU-by-task - Formal Logic | 38.89 | Accuracy (%) | 83.9 |
| MMLU-by-task - TruthfulQA MC2 | 51.56 | Accuracy (%) | 83.5 |
| MMLU-by-task - Human Sexuality | 64.12 | Accuracy (%) | 79.1 |
| MMLU-by-task - Nutrition | 61.44 | Accuracy (%) | 78.5 |
| MMLU-by-task - Econometrics | 32.46 | Accuracy (%) | 77.1 |
| MMLU-by-task - ARC Challenge | 56.31 | Accuracy (%) | 76.1 |
| MMLU-by-task - Sociology | 73.13 | Accuracy (%) | 75.3 |
| MMLU-by-task - High School Physics | 33.77 | Accuracy (%) | 75 |
| MMLU-by-task - Computer Security | 68 | Accuracy (%) | 74.9 |
| MMLU-by-task - Professional Psychology | 55.07 | Accuracy (%) | 73.4 |
| MMLU-by-task - Electrical Engineering | 48.97 | Accuracy (%) | 71.2 |
Interactive version: theaggregate.ai/model?slug=ultralm-13b-fp16 · How It Works · Data refreshed daily, snapshot 2026-11-08.