InternLM-XComposer2: benchmark results
Provider: Shanghai AI Lab. Released 2024-01-26. Access: Open.
Aggregate Rating 1386 ± 31, rank #6682 of 10916 rated models, from 445 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| OpenVLM MMT-Bench - LVLM Response Judgement | 70 | Score (%) | 99.3 |
| OpenVLM MMT-Bench - Scene Graph Recognition | 100 | Score (%) | 99.3 |
| OpenVLM MME - Artwork | 184.8 | Score | 99.1 |
| OpenVLM MME - Scene | 167.5 | Score | 97 |
| OpenVLM MMT-Bench - Geometrical Perspective | 91.7 | Score (%) | 96.6 |
| OpenVLM MMT-Bench - Body Emotion Recognition | 60 | Score (%) | 94.7 |
| OpenVLM MMT-Bench - Table Structure Recognition | 100 | Score (%) | 94.7 |
| OpenVLM MMT-Bench - Celebrity Recognition | 95 | Score (%) | 94.2 |
| OpenVLM MMT-Bench - Humanities and Social Science | 72.7 | Score (%) | 93.9 |
| OpenVLM MMT-Bench - Other Biological Attributes | 85 | Score (%) | 93.2 |
| OpenVLM MMBench V1.1 CN - Physical Property Reasoning | 77.2 | Accuracy (%) | 93.1 |
| OpenVLM MMT-Bench - One Shot Detection | 85 | Score (%) | 92.7 |
Interactive version: theaggregate.ai/model?slug=internlm-xcomposer2 · How It Works · Data refreshed daily, snapshot 2026-11-08.