ALMA-7B-R: benchmark results
Provider: Other. Access: Open.
Aggregate Rating 1210 ± 49, rank #9555 of 10916 rated models, from 13 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Portuguese LLM - FaQuAD NLI | 44.42 | Macro F1 (%) | 25.3 |
| Open Portuguese LLM - OAB Exams | 29.16 | Accuracy (%) | 24.6 |
| Open Portuguese LLM - ASSIN2 RTE | 60 | Macro F1 (%) | 20.9 |
| Open Portuguese LLM - BLUEX | 24.62 | Accuracy (%) | 19.1 |
| Open Portuguese LLM - ENEM | 22.46 | Accuracy (%) | 18.8 |
| Open Chinese LLM - WinoGrande | 59.91 | Accuracy (%) | 12.9 |
| Open Chinese LLM - HellaSwag | 48.55 | Accuracy (%) | 5.9 |
| Open Chinese LLM - ARC Challenge | 38.91 | Accuracy (%) | 5.6 |
| Open Chinese LLM - GSM8K | 4.02 | Accuracy (%) | 5 |
| Open Chinese LLM Leaderboard | 36.37 | Official Nonzero-Task Mean (%) | 3 |
| Open Chinese LLM - C-Eval Semantic | 29.64 | Accuracy (%) | 2.4 |
| Open Chinese LLM - CMMLU | 30.15 | Accuracy (%) | 2.4 |
Interactive version: theaggregate.ai/model?slug=alma-7b-r · How It Works · Data refreshed daily, snapshot 2026-11-08.