Spoken Question Answering on VoiceBench S2T
4.94AlpacaEvalLongcat-Flash-Omni-Instruct
Evaluation Results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Longcat-Flash-Omni-InstructLLM Size=560B-A27B, Frame Rate-In=12.5, Frame Rate-Out=16.672025.12 | 4.94 | 4.32 | 82.46 | 81.95 | 93.41 | 77.99 | 100 | 88.72 | |
| GPT-Audio2025.12 | 4.84 | 4.47 | 89.72 | 83.25 | 92.53 | 79.12 | 99.62 | 90.06 | |
| Fun-Audio-Chat-30B-A3BLLM Size=30B-A3B, Frame Rate-In=5, Frame Rate-Out=52025.12 | 4.82 | 4.49 | 72.87 | 75.31 | 88.57 | 77.25 | 99.23 | 85.63 | |
| Gemini-2.5-Pro2025.12 | 4.7 | 4.11 | 83.54 | 88.32 | 95.16 | 77.83 | 97.69 | 88.39 |