Math Reasoning on VoxEval Average
73.1AccuracyQwen2.5-7B-Instruct
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Qwen2.5-7B-Instructcategory=Text Topline2026.05 | 73.1 | — | |
| TextPro-SLM-7Bcategory=End-to-end Systems2026.05 | 67.8 | 5.3 | |
| Qwen2.5-Omni-7Bcategory=End-to-end Systems2026.05 | 61.6 | 11.5 | |
| Kimi-Audio-7Bcategory=End-to-end Systems2026.05 | 49.2 | 23.9 |