Virtuoso-Small-v2: benchmark results
An Arcee AI Qwen2.5-14B derivative distilled from DeepSeek-V3 logits with tokenizer alignment, fusion merging and DPO. Provider: Arcee AI. Access: Open.
Aggregate Rating 1574 ± 33, rank #1342 of 10916 rated models, from 56 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard - IFEval | 82.73 | Score | 99.2 |
| Open LLM Leaderboard - MATH Level 5 | 46.6 | Score | 94.6 |
| Open LLM Leaderboard - BBH | 65.54 | Score | 93.1 |
| Open LLM Leaderboard - MMLU-Pro | 51.88 | Score | 89.9 |
| Open LLM Leaderboard - GPQA | 35.32 | Score | 88.9 |
| Open Japanese LLM - ALT J TO E Comet Wmt22 | 88.67 | Score (%) | 84.9 |
| Open Japanese LLM - Jsts Spearman | 87.68 | Score (%) | 82.9 |
| Open Japanese LLM - MT | 84.87 | Score (%) | 82.9 |
| Open Japanese LLM - Jsick Exact Match | 83.3 | Score (%) | 82.6 |
| Open Japanese LLM - Wikicorpus E TO J Comet Wmt22 | 84.1 | Score (%) | 82 |
| Open Japanese LLM - Jsts Pearson | 90.09 | Score (%) | 81.7 |
| Open Japanese LLM - Jcommonsenseqa Exact Match | 95.71 | Score (%) | 81.6 |
Interactive version: theaggregate.ai/model?slug=virtuoso-small-v2 · How It Works · Data refreshed daily, snapshot 2026-11-08.