EMMA: Efficient Multimodal Understanding, Generation, and Editing with a Unified Architecture
Open paper
Benchmark
Model
Rank
Results
text-to-image-generation-on-geneval
EMMA (4B, original prompts)
#16
Overall: 0.91
text-to-image-generation-on-dpg-bench
EMMA
#54
DPG-Bench Score: 85.63
visual-question-answering-on-mm-vet
EMMA
–
GPT-4 score: 73.0
Rank counts only results with a code link.