BlueLM-3.5-Nano-3B: benchmark results
Provider: vivo.
Aggregate Rating 1728 ± 43, rank #239 of 10916 rated models, from 13 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| HallusionBench | 70.5 | Overall Score | 100 |
| MMStar | 84.3 | Accuracy (%) | 100 |
| MathVista | 88.3 | Accuracy (%) | 100 |
| OpenCompass Multimodal - MMBench | 89.7 | Score (%) | 100 |
| OpenCompass Multimodal - Average | 82.1 | Average Score | 98.1 |
| AI2D | 88.7 | Accuracy (%) | 95.2 |
| OpenVLM OCRBench | 90 | Score (normalized) | 94 |
| MM-Vet | 76.5 | Score | 90.4 |
| OpenCompass Multimodal - MMMU | 69.2 | Score (%) | 76.9 |
| SuperCLUE-Ondevice - Image Understanding | 91.34 | Score | 66.7 |
| SuperCLUE-Ondevice - Overall | 89.86 | Score | 66.7 |
| SuperCLUE-Ondevice - Text Understanding | 92.85 | Score | 55.6 |
Interactive version: theaggregate.ai/model?slug=bluelm-3-5-nano-3b · How It Works · Data refreshed daily, snapshot 2026-11-08.