Spoken Question Answering on UltraEval-Audio WebQ
55.95S2T Score (Task Correctness)Qwen3-Omni-30B
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Qwen3-Omni-30B2026.06 | 55.95 | 52.51 | 14.92 | |
| GPT-4o2026.06 | 50.98 | — | — | |
| VocalNet-8Bstreaming mode=true2026.06 | 44.05 | 39.27 | 7.68 | |
| Qwen2.5-Omni-7B2026.06 | 42.18 | 39.42 | 21.5 | |
| Backbone-LLM-7B2026.06 | 42.18 | — | — | |
| PRIME-Speech-9B2026.06 | 42.04 | 40.18 | 3.33 | |
| Kimi-Audio-7B2026.06 | 41.98 | 35.37 | 14.85 | |
| Step-Audio-2-Mini-7B2026.06 | 33.02 | 31.69 | 8.56 | |
| GLM-4-Voice-9B2026.06 | 32.2 | 15.9 | — |