Has anyone tried quantum randomness to drive temp functions for LLMs?

[inshard]

Would performance differ when compared to pseudo-randomness? Specifically, how does it affect deterministic outcomes and creativity, including new thinking?