Nex-N2.5-Max: benchmark results
Provider: Nex AGI. Access: Open.
Aggregate Rating 1705 ± 29, rank #322 of 10916 rated models, from 96 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Nejumi 4 - ALT - Controllability | 94.02 | Score (%) | 99.3 |
| Nejumi 4 - M-IFEval (Japanese) | 92.48 | Score (%) | 99 |
| Nejumi 4 - jaster (0-shot) - JCoLA (in-domain) | 81 | Exact match (%) | 94.8 |
| Nejumi 4 - GLP - Applied Language | 89.95 | Score (%) | 90.5 |
| Nejumi 4 - jaster (0-shot) - ALT Japanese-to-English | 90.08 | COMET wmt22 (x100) | 87.9 |
| Nejumi 4 - JTruthfulQA | 73.71 | Score (%) | 87.6 |
| Nejumi 4 - GLP - Translation | 91.3 | Score (%) | 86.5 |
| Nejumi 4 - HLE (Japanese) - Accuracy | 38.14 | Accuracy (%) | 86.2 |
| Nejumi 4 - jaster (2-shot) - JSQuAD | 80.97 | Character F1 (x100) | 85.9 |
| Nejumi 4 - jaster (0-shot) - ALT English-to-Japanese | 92.43 | COMET wmt22 (x100) | 84.1 |
| Nejumi 4 - jaster (2-shot) - ALT Japanese-to-English | 90.22 | COMET wmt22 (x100) | 84.1 |
| Nejumi 4 - jaster (0-shot) - MGSM | 92 | Exact match (%) | 83.8 |
Interactive version: theaggregate.ai/model?slug=nex-n2-5-max · How It Works · Data refreshed daily, snapshot 2026-11-08.