Blossom-V6-7B: benchmark results
A conversational fine-tune released by Azure99 based on Qwen/Qwen2.5-7B, 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 1504 ± 31, rank #2273 of 10916 rated models, from 56 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard - MATH Level 5 | 45.85 | Score | 94 |
| Open Japanese LLM - Xlsum JA Bert Score JA F1 | 70.39 | Score (%) | 85.1 |
| Open Japanese LLM - Xlsum JA RougeLsum | 25.59 | Score (%) | 83.2 |
| Open Japanese LLM - Xlsum JA Rouge1 | 29.96 | Score (%) | 81.1 |
| Open Japanese LLM - Xlsum JA Rouge2 | 11.01 | Score (%) | 80.8 |
| Open Japanese LLM - SUM | 11 | Score (%) | 80.6 |
| Open LLM Leaderboard - MMLU-Pro | 41.44 | Score | 75.5 |
| Open LLM Leaderboard - MuSR | 43.01 | Score | 70.5 |
| Open LLM Leaderboard - IFEval | 55.38 | Score | 66.9 |
| Open LLM Leaderboard - GPQA | 30.45 | Score | 61.3 |
| Open Japanese LLM - Wiki NER SET F1 | 6.19 | Score (%) | 57.2 |
| Open Japanese LLM - RC | 90.43 | Score (%) | 52.4 |
Interactive version: theaggregate.ai/model?slug=blossom-v6-7b · How It Works · Data refreshed daily, snapshot 2026-11-08.