There is a lot of evidence that geometry is closely tied with complexity in holographic models of quantum gravity. While complexity is typically hard to pin down precisely, the Python’s Lunch conjecture (PLC) makes quantitative predictions for complexity that seem strong enough to be testable. I will present evidence that the PLC is not quantitatively accurate within various tensor network models, predicting complexities that are too large in some cases and too small in others, and I will discuss the potential implications for holography.