ABEJA-Qwen2.5-32B-Japanese-v0.1: benchmark results
Provider: ABEJA. Access: Open.
Aggregate Rating 1601 ± 36, rank #1002 of 10916 rated models, from 54 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Japanese LLM - Jmmlu Exact Match | 77.72 | Score (%) | 99.5 |
| Open Japanese LLM - Jcommonsenseqa Exact Match | 97.05 | Score (%) | 99.4 |
| pfgen-bench - Completion Mode - Helpfulness | 0.55 | Helpfulness Score | 99.4 |
| Open Japanese LLM - HE | 78.85 | Score (%) | 98.3 |
| Open Japanese LLM - MC | 91.67 | Score (%) | 98.1 |
| pfgen-bench - Completion Mode - Score | 0.77 | pfgen Score (mean of three) | 98.1 |
| Open Japanese LLM - Wiki PAS SET F1 | 12 | Score (%) | 97.8 |
| Open Japanese LLM - Jemhopqa Char F1 | 66.13 | Score (%) | 97.4 |
| Open Japanese LLM - Kuci Exact Match | 85.26 | Score (%) | 96.9 |
| Open Japanese LLM - NLI | 81.56 | Score (%) | 95.2 |
| pfgen-bench - Completion Mode - Fluency | 0.85 | Fluency Score | 95.2 |
| pfgen-bench - Completion Mode - Truthfulness | 0.92 | Truthfulness Score | 94.7 |
Interactive version: theaggregate.ai/model?slug=abeja-qwen2-5-32b-japanese-v0-1 · How It Works · Data refreshed daily, snapshot 2026-11-08.