Baichuan-Omni-1.5: benchmark results
Provider: Baichuan. Released 2025-01-26.
Aggregate Rating 1483 ± 24, rank #2631 of 10916 rated models, from 115 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| StyleBench (Speaking Style) - Volume (Turn 3) | 58.13 | Valid sample percentage (%): share of the 160 volume test di | 100 |
| AVI-Bench - Audio-Referenced Visual Retrieval | 39.42 | Task score (%) on audio-referenced visual retrieval (264 que | 96 |
| AVI-Bench - Audio Sensation QA | 24.7 | Task score (%) on primitive audio sensation (502 multiple-ch | 94 |
| AEQ-Bench - Discrimination | 4.74 | Discrimination (1-6; how far the responses to one utterance | 87.5 |
| AVI-Bench - Audio-Visual Matching | 59.2 | Task score (%) on audio-visual matching (250 videos): accura | 80 |
| AVI-Bench - Understanding | 36.46 | Stage score (%) for the understanding stage: mean of the two | 80 |
| AVI-Bench - Visual-Referenced Audio Retrieval | 28.14 | Task score (%) on visual-referenced audio retrieval (264 que | 80 |
| AEQ-Bench - Naturalness | 3.88 | Naturalness (1-4; how human-like and conversational the text | 75 |
| MedMosaic - Voice QA | 31.5 | Accuracy (%; multiple choice, extracted answer letter agains | 75 |
| AVI-Bench - Primitive Sensation | 20.55 | Stage score (%) for the primitive sensation extension (AVI-B | 72 |
| AVI-Bench | 33.15 | Overall score (%): mean of the four stage averages (percepti | 68 |
| AVI-Bench - Audio Multi-Instance Classification | 22.59 | Task score (%) on audio multi-instance classification (518 c | 68 |
Interactive version: theaggregate.ai/model?slug=baichuan-omni-1-5 · How It Works · Data refreshed daily, snapshot 2026-11-08.