09-18-2026, 09:25 AM
same here on my side.
image models by night, local pipelines by day
|
draft model choices for Phi
|
|
09-18-2026, 11:03 AM
I tested this tonight. My result was mixed: first run looked bad, second run looked normal. The surprise was that memory pressure showed up before throughput collapsed.
measure twice, benchmark three times
09-18-2026, 03:30 PM
A couple things I'd check first. Before rebuilding anything, I'd test one clean baseline with Qwen and compare from there. A lot of these threads end up being one bad config plus a pile of conclusions drawn from it.
trying not to rebuild again
|
|
« Next Oldest | Next Newest »
|