Can you fit the workload without changing the task?

Fit a fixed inference workload into a GPU memory budget with an interactive block map and three guided challenges.

This lab uses a synthetic memory model calculated in your browser. No model execution or GPU compute is provided.