Awaking Spatial Intelligence in Unified Multimodal Understanding and Generation

Benchmark Model Rank Results
image-editing-on-gedit-bench-enJoyAI-Image-Edit (w/ PE)#3Overall: 8.290
text-to-image-generation-on-longtext-bench-zhJoyAI-Image#6Accuracy: 96.3
text-to-image-generation-on-longtext-benchJoyAI-Image#7Accuracy: 96.3
visual-question-answering-vqa-on-blinkJoyAI-Image-Und#7Accuracy: 69.6
text-to-image-generation-on-cvtg-2kJoyAI-Image#10Word Accuracy: 0.8739NED: 0.9369
text-to-image-generation-on-oneig-enJoyAI-Image#10OneIG Overall Score: 0.542
text-to-image-generation-on-oneig-zhJoyAI-Image#11OneIG Overall Score: 0.521
text-to-image-generation-on-dpg-benchJoyAI-Image#15DPG-Bench Score: 88.05
visual-question-answering-vqa-on-realworldqaJoyAI-Image-Und#29Accuracy: 74.0
vision-language-understanding-on-mmstarJoyAI-Image-Und#35Accuracy: 71.3
multimodal-reasoning-on-mathvistaJoyAI-Image-Und#50Accuracy: 74.4