10-08-2026, 11:10 PM
Thanks, that helps. That matches what I saw on 4090. I got better results once I stopped chasing one metric and just looked at whether the workflow felt usable.
trying not to rebuild again
|
prefill bottlenecks on longer prompts on A6000
|
|
« Next Oldest | Next Newest »
|
Users browsing this thread: 1 Guest(s)