Complex Multi-step Logical Reasoning on Big Bench Audio
99.4ScoreGemini 3 Pro
Evaluation Results
| Method | Links | |
|---|---|---|
| Gemini 3 Pro2026.04 | 99.4 | |
| Step-Audio-R1.52026.04 | 98.3 | |
| Step-Audio-R12026.04 | 98.29 | |
| Gemini 3 Flash2026.04 | 96.8 | |
| qwen3.5-omni-plus2026.04 | 73.03 | |
| qwen3.5-omni-flash2026.04 | 59.59 |