Ultiima-32B: benchmark results
Provider: Sakalti. Access: Open.
Aggregate Rating 1590 ± 36, rank #1116 of 10916 rated models, from 56 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard - MMLU-Pro | 59.1 | Score | 99.7 |
| Open LLM Leaderboard - BBH | 70.37 | Score | 99.1 |
| Open LLM Leaderboard - MuSR | 49.95 | Score | 98.3 |
| Open Japanese LLM - Wiki PAS SET F1 | 11.81 | Score (%) | 97.1 |
| Open LLM Leaderboard - MATH Level 5 | 49.62 | Score | 96.6 |
| Open Japanese LLM - Jsick Exact Match | 86 | Score (%) | 96.4 |
| Open Japanese LLM - Jnli Exact Match | 89.69 | Score (%) | 95.1 |
| Open LLM Leaderboard - GPQA | 38.09 | Score | 95.1 |
| Open Japanese LLM - MR | 94.2 | Score (%) | 94 |
| Open Japanese LLM - Jmmlu Exact Match | 75.32 | Score (%) | 92.3 |
| Open Japanese LLM - HE | 77.69 | Score (%) | 92.2 |
| Open Japanese LLM - NLI | 80.92 | Score (%) | 90.3 |
Interactive version: theaggregate.ai/model?slug=ultiima-32b · How It Works · Data refreshed daily, snapshot 2026-11-08.