SmolVLM-500M: benchmark results
Provider: HuggingFace. Released 2025-01-23. Access: Open.
Aggregate Rating 1238 ± 29, rank #9356 of 10916 rated models, from 499 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| OpenVLM MMT-Bench - 3D Indoor Recognition | 60 | Score (%) | 99.5 |
| OpenVLM MMMU - Music | 46.7 | Accuracy (%) | 96.8 |
| OpenVLM MMT-Bench - GUI Install | 60 | Score (%) | 94.9 |
| OpenVLM MMT-Bench - Action Quality Assessment | 35 | Score (%) | 94.2 |
| OpenVLM MMT-Bench - Color Constancy | 71.4 | Score (%) | 94.2 |
| OpenVLM MMT-Bench - 3D | 55 | Score (%) | 90.8 |
| OpenVLM MMT-Bench - Image Matting | 30 | Score (%) | 90.3 |
| OpenVLM MMT-Bench - Sketch-to-Code | 40 | Score (%) | 85.7 |
| OpenVLM MMT-Bench - Multiple View Image Understanding | 35 | Score (%) | 82.3 |
| OpenVLM MMT-Bench - Animal Keypoint Detection | 40 | Score (%) | 82 |
| OpenVLM MMT-Bench - Spot The Diff | 65 | Score (%) | 77.9 |
| OpenVLM MMT-Bench - Raven Progressive Matrices | 20 | Score (%) | 76.2 |
Interactive version: theaggregate.ai/model?slug=smolvlm-500m · How It Works · Data refreshed daily, snapshot 2026-11-08.