InternVL3-9B: benchmark results
Provider: Shanghai AI Lab. Released 2025-04-11. Access: Open.
Aggregate Rating 1546 ± 13, rank #1756 of 10916 rated models, from 584 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| OpenVLM CCBench - Cultural Relics | 93.8 | Accuracy (%) | 99.6 |
| OpenVLM MMT-Bench - Vehicle Retrieval | 95 | Score (%) | 99.5 |
| OpenVLM SEEDBench IMG - Scene Understanding | 80.9 | Accuracy (%) | 99.1 |
| OpenVLM MMT-Bench - Small Object Detection | 80 | Score (%) | 99 |
| OpenVLM MMT-Bench - Attribute Hallucination | 90 | Score (%) | 98.8 |
| OpenVLM MMT-Bench - Chart To Text | 100 | Score (%) | 98.8 |
| OpenVLM MMT-Bench - Chart VQA | 85 | Score (%) | 98.8 |
| OpenVLM MMT-Bench - Image Quality Assessment | 75 | Score (%) | 98.5 |
| OpenVLM MMStar - Coarse Perception | 75.2 | Accuracy (%) | 98.2 |
| OpenVLM MMBench V1.1 CN - Social Relation | 91.2 | Accuracy (%) | 98.1 |
| OpenVLM MMT-Bench - Text-to-Image Retrieval | 75 | Score (%) | 98.1 |
| OpenVLM MMT-Bench - Handwritten Mathematical Expression Recognition | 100 | Score (%) | 97.8 |
Interactive version: theaggregate.ai/model?slug=internvl3-9b · How It Works · Data refreshed daily, snapshot 2026-11-08.