VisualGLM: benchmark results
Provider: Zhipu. Released 2023-05-17. Access: Open.
Aggregate Rating 1138 ± 48, rank #10152 of 10916 rated models, from 406 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| OpenVLM MMT-Bench - Image Matting | 40 | Score (%) | 97.8 |
| OpenVLM MMT-Bench - Temporal Ordering | 40 | Score (%) | 92 |
| OpenVLM MMT-Bench - Navigation | 80 | Score (%) | 82 |
| OpenVLM MMT-Bench - Science | 41.7 | Score (%) | 78.6 |
| OpenVLM MMT-Bench - Point Tracking | 70 | Score (%) | 75.2 |
| OpenVLM MMT-Bench - Spot The Diff | 55 | Score (%) | 73.8 |
| OpenVLM MMT-Bench - Geometrical Relativity | 60 | Score (%) | 72.1 |
| OpenVLM MMT-Bench - Multiple View Image Understanding | 30 | Score (%) | 71.1 |
| OpenVLM MMT-Bench - Multiple Image Captioning | 90 | Score (%) | 67 |
| OpenVLM MMMU - Music | 33.3 | Accuracy (%) | 63.6 |
| OpenVLM MMT-Bench - MeViS | 35 | Score (%) | 62.9 |
| OpenVLM MMT-Bench - Landmark Recognition | 100 | Score (%) | 57.3 |
Interactive version: theaggregate.ai/model?slug=visualglm · How It Works · Data refreshed daily, snapshot 2026-11-08.