| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Multimodal Understanding | Image Understanding Benchmarks MMB(EN), MMB(ZH), POPE, MMStar, MMMUVal, AI2D, MuirBench, BLINK, RealWorld | MMB (EN)86.08 | 21 | |
| Image Understanding | Image Understanding benchmarks GQA, MME, POPE, VQA^T, MMB, SQA | GQA Score62.2 | 10 |