Ultiima-14B: benchmark results
Provider: Sakalti. Access: Open.
Aggregate Rating 1566 ± 34, rank #1459 of 10916 rated models, from 56 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard - MMLU-Pro | 53.81 | Score | 95.1 |
| Open LLM Leaderboard - MATH Level 5 | 46.98 | Score | 94.8 |
| Open LLM Leaderboard - GPQA | 37.42 | Score | 93.5 |
| Open Japanese LLM - Wiki Coreference SET F1 | 9.56 | Score (%) | 93.1 |
| Open LLM Leaderboard - MuSR | 47.18 | Score | 92 |
| Open LLM Leaderboard - BBH | 64.91 | Score | 90.8 |
| Open Japanese LLM - Wiki PAS SET F1 | 8.17 | Score (%) | 78.2 |
| Open Japanese LLM - Janli Exact Match | 82.78 | Score (%) | 77.5 |
| Open Japanese LLM - Jcommonsenseqa Exact Match | 95.35 | Score (%) | 71.6 |
| Open Japanese LLM - Niilc Char F1 | 53.1 | Score (%) | 70.8 |
| Open Japanese LLM - Jemhopqa Char F1 | 57.5 | Score (%) | 70.3 |
| Open Japanese LLM - Xlsum JA Bleu JA | 2.83 | Score (%) | 70.1 |
Interactive version: theaggregate.ai/model?slug=ultiima-14b · How It Works · Data refreshed daily, snapshot 2026-11-08.