Add Moondream VLM Architecture #2549
Description
Vision Encoder: SigLIP (Sigmoid Loss for Language Image Pre-Training).
Text Decoder: Phi-1.5 (Microsoft's tiny model).
Connector: simple Projection Layer.
{{ message }}
Vision Encoder: SigLIP (Sigmoid Loss for Language Image Pre-Training).
Text Decoder: Phi-1.5 (Microsoft's tiny model).
Connector: simple Projection Layer.