| Benchmark | Model | Rank | Results |
|---|---|---|---|
| visual-question-answering-vqa-on-seed-seedbench | RAE (SigLIP-2) | #24 | Accuracy: 69.8 |
| text-to-image-generation-on-dpg-bench | RAE (SigLIP-2 ViT-So) | #78 | DPG-Bench Score: 76.9 |
| multimodal-reasoning-on-mmmu | RAE (SigLIP-2) | #123 | Accuracy: 41.1 |
| text-to-image-generation-on-geneval | RAE (Qwen-2.5 7B + DiT 5.5B, without TTS) | #156 | Overall: 0.555 |