L3.1-Niitorm-8B-DPO-t0.0001: benchmark results
v000000's NEARSWAP merge of Niitama 1.1 and Llama-3.1-Storm 8B, DPO-trained on Gutenberg fiction for more human-sounding prose. Provider: v000000. Released 2024-09-19. Access: Open.
Aggregate Rating 1488 ± 21, rank #823 of 1654 rated models, from 20 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-GPQA-Diamond | 29.29 | Normalized accuracy (%) | 99.2 |
| Open Korean LLM - Ko-IFEval | 52.87 | Strict accuracy, prompt/instruction mean (%) | 95.3 |
| Open Korean LLM Leaderboard | 47 | Average Score (%) | 94.4 |
| Open LLM Leaderboard - IFEval | 76.89 | Score | 92.8 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 58 | Exact match, flexible extract (%) | 92.1 |
| Open Korean LLM - Ko-EQ-Bench | 35.08 | EQ-Bench score | 85 |
| Open Portuguese LLM - BLUEX | 60.78 | Accuracy (%) | 79.2 |
| Open Portuguese LLM - OAB Exams | 51.07 | Accuracy (%) | 79.1 |
| Open Portuguese LLM - ENEM | 70.89 | Accuracy (%) | 78 |
| Open Portuguese LLM - ASSIN2 RTE | 92.15 | Macro F1 (%) | 74.9 |
| Open Portuguese LLM - FaQuAD NLI | 75.68 | Macro F1 (%) | 74.3 |
| Open LLM Leaderboard - MMLU-Pro | 38.66 | Score | 69 |
Interactive version: theaggregate.ai/model?slug=l3-1-niitorm-8b-dpo-t0-0001 · How It Works · Data refreshed daily, snapshot 2026-10-24.