batching changes after update

[audioriver]

Thought I'd post a small recipe because a lot of threads ask similar things.

This isn't revolutionary, but the combination of PyTorch and M3 Max felt surprisingly usable once I tuned around prefill/decode split.

If someone wants the exact settings I can clean them up and post them.

For context, I'm not chasing leaderboard numbers. I mostly want the system to feel consistent day to day.