BlueLM-2.6-3B: benchmark results
Provider: vivo.
Aggregate Rating 1665 ± 37, rank #526 of 10916 rated models, from 145 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| OpenVLM MMStar - Logical Reasoning | 85.6 | Accuracy (%) | 100 |
| OpenVLM MathVista - ARI | 77.1 | Accuracy (%) | 100 |
| OpenVLM MathVista - MWP | 95.2 | Accuracy (%) | 100 |
| OpenVLM MathVista - STA | 89.7 | Accuracy (%) | 99.8 |
| OpenVLM MMBench V1.1 CN - CP | 86.5 | Accuracy (%) | 99.6 |
| OpenVLM MMBench V1.1 CN - Image Quality | 79.8 | Accuracy (%) | 99.6 |
| OpenVLM MMBench V1.1 EN - Image Quality | 83.1 | Accuracy (%) | 99.6 |
| OpenVLM MMStar - Fine-Grained Perception | 70.8 | Accuracy (%) | 99.6 |
| OpenVLM MMStar - Instance Reasoning | 86.8 | Accuracy (%) | 99.6 |
| OpenVLM MMStar - Coarse Perception | 81.6 | Accuracy (%) | 99.3 |
| OpenVLM MMBench V1.1 EN - CP | 86.3 | Accuracy (%) | 99.1 |
| OpenVLM MathVista - GPS | 92.8 | Accuracy (%) | 99.1 |
Interactive version: theaggregate.ai/model?slug=bluelm-2-6-3b · How It Works · Data refreshed daily, snapshot 2026-11-08.