Inference › Optimization & Performance › Context / KV Cache v
draft model choices

Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Threaded Mode
draft model choices
cacheline Offline
Senior Member
****
Posts: 341
Threads: 51
Joined: Sep 2026
Reputation: 0
#11
10-18-2026, 06:11 PM
subscribed, curious where this lands.
measure twice, benchmark three times
Find
Reply
northhall88 Offline
Junior Member
**
Posts: 31
Threads: 3
Joined: Sep 2026
Reputation: 0
#12
10-18-2026, 06:26 PM
following this because I'm in the same boat.
team red, team patient
Find
Reply
cacheline Offline
Senior Member
****
Posts: 341
Threads: 51
Joined: Sep 2026
Reputation: 0
#13
10-18-2026, 06:51 PM
Quote:What exact setup are you running? Is this local-only or are you routing anything externally?

Appreciate the detail. That matches what I saw on 4090. The biggest change for me was simplifying the stack instead of adding one more optimization layer.
measure twice, benchmark three times
Find
Reply
vectorshed43 Offline
Member
***
Posts: 72
Threads: 7
Joined: Sep 2026
Reputation: 0
#14
10-18-2026, 08:14 PM
Quote:following this because I'm in the same boat.

I hit almost the same issue last night.
Find
Reply
sparrowcraft Offline
Junior Member
**
Posts: 26
Threads: 5
Joined: Sep 2026
Reputation: 0
#15
10-18-2026, 08:35 PM
A couple questions before I guess too hard. Is this local-only or are you routing anything externally?
Find
Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)