Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
MCP server discovery mess on GB200
#11
worth checking the driver first.
queue depth watcher
Reply
#12
Quote:worth checking the driver first.

My first guess would be this: Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at batch size. Not glamorous advice, but repeatable baselines save a lot of time. Turns out 'just one more test' was a lie.
Reply
#13
That matches what I was seeing. I don't think your conclusion is far off. Worth rechecking your defaults though, because one bad assumption can make the rest of the thread look confusing.
Reply
#14
Interesting, I hadn't tried that. I don't think your conclusion is far off.
Reply
#15
Appreciate the detail. I ended up in about the same place using MLX. The biggest change for me was simplifying the stack instead of adding one more optimization layer.
Reply
#16
A couple things I'd check first. Before rebuilding anything, I'd test one clean baseline with Qwen and compare from there.
learning in public
Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)