llamaRAGdrama: benchmark results
kevin009's llamaRAGdrama, a Llama fine-tune for question-answering and retrieval-augmented generation, trained on dramatic text data enriched with factual sources to stay accurate under stylized prompts. Provider: kevin009. Released 2024-02-04. Access: Open.
Aggregate Rating 1469 ± 30, rank #878 of 1635 rated models, from 17 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - WinoGrande | 86.66 | Accuracy (%) (5-shot) | 99.3 |
| Open LLM Leaderboard v1 - HellaSwag | 88.83 | Normalized accuracy (%) (10-shot) | 94.3 |
| Open LLM Leaderboard v1 - ARC Challenge | 72.01 | Normalized accuracy (%) (25-shot) | 90.7 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 70.24 | MC2 (%) (0-shot) | 90.2 |
| Open LLM Leaderboard v1 - GSM8K | 65.66 | Accuracy (%) (5-shot) | 80.4 |
| Open Portuguese LLM - ASSIN2 RTE | 92.24 | Macro F1 (%) | 76.6 |
| Open LLM Leaderboard - MuSR | 43.16 | Score | 71.7 |
| Open LLM Leaderboard v1 - MMLU | 64.5 | Accuracy (%) (5-shot) | 71.5 |
| Open Portuguese LLM - FaQuAD NLI | 73.42 | Macro F1 (%) | 69 |
| Open Portuguese LLM - BLUEX | 54.66 | Accuracy (%) | 60.8 |
| Open Portuguese LLM - ENEM | 62.98 | Accuracy (%) | 54.1 |
| Open Portuguese LLM - OAB Exams | 42.73 | Accuracy (%) | 53.1 |
Interactive version: theaggregate.ai/model?slug=llamaragdrama · How It Works · Data refreshed daily, snapshot 2026-11-05.