LiteVLM: A Low-Latency Vision-Language Model Inference Pipeline for Resource-Constrained Environments

VLMRe-assign community