Multi-turn Interaction on Audio Multi-Challenge Text Modality
49.21Average Pass Rate (APR)Gemini-3-Flash-Preview
Evaluation Results
| Method | Links | |
|---|---|---|
| Gemini-3-Flash-PreviewModel Series=Gemini-3-Flash-Preview2026.03 | 49.21 | |
| Gemini-2.5-FlashModel Series=Gemini-2.5-Flash2026.03 | 43.22 | |
| GPT-4oModel Series=GPT-4o, Modality=Base, Text-modality2026.03 | 32.13 | |
| Qwen3-A3B-InstructModel Series=Qwen3-A3B-Instruct, Modality=Base, Text-modality2026.03 | 29.31 | |
| X-OPDModel Series=Omni-A3B-Instruct, Training Paradigm=X-OPD2026.03 | 28.66 | |
| Omni-A3B-InstructModel Series=Omni-A3B-Instruct, Training Paradigm=Base2026.03 | 26.14 | |
| Ministral-3-3BModel Series=Ministral-3-3B, Modality=Base, Text-modality2026.03 | 22.67 | |
| Omni-A3B-Instruct + Offline KDModel Series=Omni-A3B-Instruct, Training Paradigm=Offline KD2026.03 | 21.27 | |
| Omni-A3B-Instruct + GKD (Forward KL)Model Series=Omni-A3B-Instruct, Training Paradigm=GKD (Forward KL)2026.03 | 20.76 | |
| Omni-A3B-Instruct + SFTModel Series=Omni-A3B-Instruct, Training Paradigm=SFT2026.03 | 19.25 | |
| Voxtral-Mini-3BModel Series=Voxtral-Mini-3B2026.03 | 17.27 | |
| Qwen2.5-7B-InstructModel Series=Qwen2.5-7B-Instruct, Modality=Base, Text-modality2026.03 | 17 | |
| Omni-7B-InstructModel Series=Omni-7B-Instruct2026.03 | 12.66 |