[pim] Re: LLM?

[[e-mail]]

Hi Jeff, 
Yes. You are right! 
We focus more on the sparse mode, such as MoE.
Sandy




发自我的zMail



 




Original
 




From:jefftant.ietf<[e-mail]>


To:张征00007940;


Cc:James.A.Stevens=[e-mail];pim<[e-mail]>;


Date:2025-07-22 14:43:52


Subject:Re: [pim] Re: LLM?








Sandy,
Perhaps you should be more specific, there are LLMs that are based on dense transformers architecture where multicast on not really applicable. 



Cheers,
Jeff


On Jul 21, 2025, at 23:27, [e-mail] wrote:











Yes. Large language model.
Sandy







 
 

_______________________________________________
pim mailing list -- [e-mail]
To unsubscribe send an email to [e-mail]













From:Stevens,Jim<James.A.Stevens=[e-mail]>


To:pim<[e-mail]>;


Date:2025-07-21 16:43:30


Subject:[pim] Re: LLM?






_______________________________________________
pim mailing list -- [e-mail]
To unsubscribe send an email to [e-mail]

Date: Fri, 18 Jul 2025 18:10:43 +0800 (CST)
From: <[e-mail]>
Subject: [pim]  Invitation to the "Multicast in LLM" side meeting
   (Tuesday July 22, 9:45 - 11:00 in room El Escorial)
To: <[e-mail]>
 
Hi, 
If you are familiar with LLM, or are interested in network layer multicast, and want to use network layer multicast to assist in implementing LLM's multicast operations, please come to the side meeting to discuss.
 
What do you mean by LLM?
·         Large Language Model AI?  (my assumption, but want to double check)
·         Low Latency Messaging?
·         Or?
 
Thanks,
Sandy