Blossom-V6-32B: benchmark results
A conversational fine-tune released by Azure99 based on Qwen/Qwen2.5-32B, using Azure99/blossom-v6-sft-stage1, Azure99/blossom-v6-sft-stage2. The card specifies one epoch for stage 1 and three epochs for stage 2. Provider: Azure99. Access: Open.
Aggregate Rating 1566 ± 36, rank #1454 of 10916 rated models, from 50 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Japanese LLM - Jsts Spearman | 88.62 | Score (%) | 97.8 |
| Open Japanese LLM - Jcommonsenseqa Exact Match | 96.51 | Score (%) | 97.5 |
| Open Japanese LLM - Wiki NER SET F1 | 15.93 | Score (%) | 96.9 |
| Open Japanese LLM - Jsts Pearson | 90.6 | Score (%) | 95 |
| Open Japanese LLM - Jsem Exact Match | 80.62 | Score (%) | 94.2 |
| Open Japanese LLM - RC | 92.14 | Score (%) | 91.1 |
| Open Japanese LLM - Wiki Dependency SET F1 | 42.91 | Score (%) | 90.6 |
| Open Japanese LLM - FA | 30.08 | Score (%) | 90.1 |
| Open Japanese LLM - Commonsensemoralja Exact Match | 91.11 | Score (%) | 88.1 |
| Open Japanese LLM - ALT J TO E Comet Wmt22 | 88.75 | Score (%) | 88 |
| Open Japanese LLM - ALT E TO J Comet Wmt22 | 90.99 | Score (%) | 87.5 |
| Open Japanese LLM - MT | 85.03 | Score (%) | 87.5 |
Interactive version: theaggregate.ai/model?slug=blossom-v6-32b · How It Works · Data refreshed daily, snapshot 2026-11-08.