Have you encountered a problem where a 1 GB transformer-based model increases even up to 8 GB when deployed using Docker containerization?
Reducing the Size of Docker Images Serving LLM Models
Search