InternVL2.5-8B-MPO: benchmark results
Provider: Shanghai AI Lab. Released 2024-12-20. Access: Open.
Aggregate Rating 1524 ± 23, rank #2022 of 10916 rated models, from 458 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AgentRx (Patient Summary, Zero-Shot) - Length of Stay | 0.62 | AUROC (0-1; stay beyond seven days; MIMIC-IV ICU test split; | 100 |
| OpenVLM CCBench - Food and Clothes | 88.7 | Accuracy (%) | 99.6 |
| CCBench | 82.9 | Accuracy (%) | 98.7 |
| OpenVLM OCRBench - Doc-oriented VQA | 92 | Correct Answers (%) | 97.9 |
| OpenVLM A-Bench - Aesthetic Quality | 70.3 | Accuracy (%) | 97.8 |
| OpenVLM MMT-Bench - Image-to-Image Retrieval | 85 | Score (%) | 97.3 |
| OpenVLM MME - Text Translation | 200 | Score | 97 |
| OpenVLM MMBench V1.1 CN - Image Scene | 92.4 | Accuracy (%) | 96.6 |
| OpenVLM MMT-Bench - HLN | 77.5 | Score (%) | 96.6 |
| OpenVLM MMT-Bench - Order Hallucination | 60 | Score (%) | 96.6 |
| OpenVLM CCBench - Cultural Relics | 87.6 | Accuracy (%) | 96.5 |
| OpenVLM MMBench V1.1 CN - Attribute Recognition | 96.6 | Accuracy (%) | 96.5 |
Interactive version: theaggregate.ai/model?slug=internvl2-5-8b-mpo · How It Works · Data refreshed daily, snapshot 2026-11-08.