PR #25619 Update MoE and qMoE spec

Update MoE and qMoE spec #25619

tianleiwu merged 16 commits into main from tlwu/moe_spec

update moe spec

e7cb8448

tianleiwu marked this pull request as draft 207 days ago

github-actions commented on 2025-08-01

github-advanced-security commented on 2025-08-01

update doc

bd36de44

github-actions commented on 2025-08-01

format

1d70f69d

add swiglu limit

451814f9

Merge branch 'main' into tlwu/moe_spec

fa0224cd

CPU change from apsonawane

4a0d84f4

use moe_helper in CPU

03a61467

remove MoEQuantType

6a5871e8

Fix build

c4eb332a

Add swiglu parameters

08c31146

Merge branch 'main' into tlwu/moe_spec

b7de4a7c

update doc

edd065cb

improve backward compatible

1b72088f

tianleiwu marked this pull request as ready for review 206 days ago

Revert "emsdk" change

b1562ddc

refacotring

37abf5d5

Disable cpu qmoe test

0635f113

kunal-vaishnavi commented on 2025-08-02

kunal-vaishnavi approved these changes on 2025-08-02

tianleiwu merged 562760a5 into main 206 days ago

tianleiwu deleted the tlwu/moe_spec branch 206 days ago

jywu-msft added release:1.23.0

tianleiwu added cherry-picked

tianleiwu removed release:1.23.0

Reviewers

kunal-vaishnavi

github-actions

github-advanced-security

Assignees

No one assigned

Labels

cherry-picked

Milestone

No milestone

onnxruntime Update MoE and qMoE spec #25619 Merged

Update MoE and qMoE spec #25619

onnxruntime
Update MoE and qMoE spec
#25619

Merged