Luminous Supreme (70B): benchmark results
Provider: Aleph Alpha. Released 2022-08-15. Access: API.
Aggregate Rating 1276 ± 26, rank #9012 of 10916 rated models, from 56 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| HELM Classic - IMDB | 95.93 | Exact Match (%) | 90.9 |
| HELM Classic - Dyck | 72.87 | Exact Match (%) | 89.7 |
| HELM Classic - MATH | 14.92 | Equivalent (%) | 76.5 |
| HELM Classic - BBQ | 41.1 | Exact Match (%) | 75.6 |
| HELM Classic - XSUM | 13.56 | ROUGE-2 (%) | 75.6 |
| HELM Classic - Synthetic Reasoning Abstract | 31.22 | Exact Match (%) | 75 |
| HELM Classic - WikiFact | 33.45 | Exact Match (%) | 74.2 |
| HELM Classic - NarrativeQA | 71.1 | F1 (%) | 73.8 |
| HELM Classic - CNN/DailyMail | 14.98 | ROUGE-2 (%) | 73.2 |
| HELM Classic - NaturalQuestions Open Book | 64.86 | F1 (%) | 72.3 |
| HELM Lite - LegalBench - International Citizenship Questions | 59 | Exact Match (%) | 68.3 |
| HELM Classic - CivilComments | 56.17 | Exact Match (%) | 65.2 |
Interactive version: theaggregate.ai/model?slug=luminous-supreme-70b · How It Works · Data refreshed daily, snapshot 2026-11-08.