Ovis1.5-Gemma2-9B: benchmark results
Provider: Alibaba. Released 2024-07-30. Access: Open.
Aggregate Rating 1458 ± 25, rank #3437 of 10916 rated models, from 441 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| OpenVLM Q-Bench - Yes-or-No (Others) | 81.7 | Accuracy (%) | 99.4 |
| OpenVLM MMT-Bench - Social Relation Recognition | 80 | Score (%) | 99.3 |
| OpenVLM MMT-Bench - Image-to-Image Retrieval | 85 | Score (%) | 97.3 |
| OpenVLM MMT-Bench - Interactive Segmentation | 78.6 | Score (%) | 97.1 |
| OpenVLM MMMU - Music | 46.7 | Accuracy (%) | 96.8 |
| OpenVLM AI2D - Food Chains and Webs | 96.2 | Accuracy (%) | 96.3 |
| OpenVLM MMT-Bench - RR | 80 | Score (%) | 95.6 |
| OpenVLM MMT-Bench - Counting By Visual Prompting | 50 | Score (%) | 95.4 |
| OpenVLM MMT-Bench - Meme Video Understanding | 85 | Score (%) | 95.4 |
| OpenVLM MMT-Bench - Texture Material Recognition | 85 | Score (%) | 95.4 |
| OpenVLM MathVista - MWP | 87.1 | Accuracy (%) | 95.4 |
| OpenVLM MMT-Bench - Table Structure Recognition | 100 | Score (%) | 94.7 |
Interactive version: theaggregate.ai/model?slug=ovis1-5-gemma2-9b · How It Works · Data refreshed daily, snapshot 2026-11-08.