InternVL2.5: benchmark results
Provider: Shanghai AI Lab.
Aggregate Rating 1578 ± 35, rank #1274 of 10916 rated models, from 20 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Video-MME - Short Videos (w/o subs) | 82.8 | Accuracy (%) | 98 |
| Video-MME - Short Videos (w/ subs) | 83.2 | Accuracy (%) | 90 |
| Video-MME - Medium Videos (w/o subs) | 70.9 | Accuracy (%) | 88 |
| Video-MME - Long Videos (w/o subs) | 62.6 | Accuracy (%) | 86 |
| Physical Reasoning - MVPBench | 39.9 | Accuracy (%) | 83.3 |
| Video-MME - Medium Videos (w/ subs) | 74.1 | Accuracy (%) | 80 |
| VideoMME w sub. | 74 | Score (imported) | 73.6 |
| Video-MME - Long Videos (w/ subs) | 64.8 | Accuracy (%) | 72 |
| V-STaR - Medium mLGM | 23.06 | Score | 61.5 |
| V-STaR | 21.26 | Average All Score | 53.8 |
| V-STaR - All mLGM | 24.92 | Score | 53.8 |
| V-STaR - All mAM | 17.6 | Score | 46.2 |
Interactive version: theaggregate.ai/model?slug=internvl2-5 · How It Works · Data refreshed daily, snapshot 2026-11-08.