DeepSeek-VL2

model
DeepSeek
MoE vision-language model for advanced multimodal understanding.
Paper (arXiv)GitHubHuggingFace

Paper

Citations 22
arXiv HTML
multimodalopen-weightmoe

Related