I am sorry I did not understand all of it. But, would this allow running large MoE on a local network with experts spread out over multiple cheaper GPUs (or even CPUs)? This would perhaps be more useful than over the Internet, within offices for example.
P2P inference sounds really nice and it would bring us back to the 2000s culture if not for the fact inference is already so cheap. Even so it is an insurance policy if the cloud providers or governments have ideas of further censoring and monitoring chats.
Ho thanks for the comment.
Verification does not depend on temperature.
Expert execution is deterministic (pure matmul).
LUMABRI_VERIFY=N re-runs N% of the calls on a second replica and requires byte-identical output.
Temperature (and sampling) happens only on the chatter, after the experts return their activations.
So it can be any value (0, 0.7, 1.2…) without affecting the verification contract.
J/k, this looks cool :)