prefill/decode separation in practice

Thread Rating:

0 Vote(s) - 0 Average
1
2
3
4
5

Linear Mode

prefill/decode separation in practice

batchsize Offline

Offline

Member

Posts: 137
Threads: 16
Joined: Sep 2026
Reputation: 0

#3

09-04-2026, 07:26 PM

Can you fill in a bit more context? Are you testing this on dual 3090 or something else?

throughput matters, but tail latency matters too

Find

Reply

« Next Oldest | Next Newest »

Messages In This Thread

prefill/decode separation in practice - by mangovault67 - 09-04-2026, 12:20 PM

Re: prefill/decode separation in practice - by batchcraft - 09-04-2026, 04:54 PM

Re: prefill/decode separation in practice - by batchsize - 09-04-2026, 07:26 PM

Re: prefill/decode separation in practice - by thetashed24 - 09-06-2026, 10:51 AM

Re: prefill/decode separation in practice - by rocmrunner40 - 09-04-2026, 10:03 PM

Re: prefill/decode separation in practice - by oxidemixer - 09-08-2026, 03:10 AM

Re: prefill/decode separation in practice - by mangofield33 - 09-06-2026, 03:24 PM

Re: prefill/decode separation in practice - by promptpoint - 09-06-2026, 11:50 PM

Re: prefill/decode separation in practice - by thetagrid15 - 09-07-2026, 12:31 AM

Re: prefill/decode separation in practice - by thetaplane - 09-08-2026, 01:21 AM

Re: prefill/decode separation in practice - by mangofield33 - 09-08-2026, 11:57 AM

Re: prefill/decode separation in practice - by cudapoint - 09-09-2026, 12:40 PM

Re: prefill/decode separation in practice - by oxidemixer - 09-12-2026, 01:17 PM

Re: prefill/decode separation in practice - by audioforge - 09-12-2026, 02:14 PM

Re: prefill/decode separation in practice - by cacheline - 09-12-2026, 03:39 PM

View a Printable Version

Forum Jump:

Users browsing this thread: 1 Guest(s)