InternLM-XComposer2.5: benchmark results
Provider: Shanghai AI Lab. Released 2024-07-03. Access: Open.
Aggregate Rating 1438 ± 20, rank #4293 of 10916 rated models, from 540 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MVPBench - Video Brightness Evaluation | 37.32 | MVPBench normalized proficiency score (accuracy minus chance | 100 |
| OpenVLM MME - Artwork | 187.2 | Score | 99.8 |
| OpenVLM MMT-Bench - Relation Hallucination | 90 | Score (%) | 99.5 |
| OpenVLM MMT-Bench - Traffic Participants Understanding | 75 | Score (%) | 99.3 |
| OpenVLM MMT-Bench - Salient Object Detection RGBD | 70 | Score (%) | 99 |
| OpenVLM MMT-Bench - Sports Recognition | 100 | Score (%) | 98.5 |
| OpenVLM MMMU - Literature | 93.3 | Accuracy (%) | 98.1 |
| OpenVLM MMT-Bench - Helmet Anomaly Detection | 95 | Score (%) | 97.1 |
| OpenVLM MMT-Bench - Traffic Light Understanding | 100 | Score (%) | 96.4 |
| OpenVLM MMT-Bench - Meme Video Understanding | 85 | Score (%) | 95.4 |
| OpenVLM MMT-Bench - Pixel Recognition | 75 | Score (%) | 94.7 |
| OpenVLM MMT-Bench - Other Biological Attributes | 85 | Score (%) | 93.2 |
Interactive version: theaggregate.ai/model?slug=internlm-xcomposer2-5 · How It Works · Data refreshed daily, snapshot 2026-11-08.