TokenPacker: Efficient Visual Projector for Multimodal LLM
Open paper
Benchmark
Model
Rank
Results
visual-question-answering-on-mm-vet
LLaVA-TokenPacker (Vicuna-13B)
#147
GPT-4 score: 34.1
visual-question-answering-on-mm-vet
LLaVA-TokenPacker (Vicuna-7B)
#182
GPT-4 score: 29.6
Rank counts only results with a code link.