PR #25558 Add support for QMoE in CPU

Add support for QMoE in CPU #25558

tianleiwu merged 20 commits into main from asonawane/qmoe

apsonawane requested a review from

tianleiwu 197 days ago

github-advanced-security commented on 2025-07-28

apsonawane changed the title ~~Asonawane/qmoe~~ Add support for QMoE in CPU 197 days ago

github-actions commented on 2025-07-28

apsonawane force pushed from fb8c2aed to 6a36b9af 196 days ago

tianleiwu commented on 2025-07-28

apsonawane force pushed from 6a36b9af to deb866a7 196 days ago

apsonawane force pushed from deb866a7 to 67ad2f4e 196 days ago

apsonawane force pushed from b0fe68a6 to 731021d1 196 days ago

tianleiwu commented on 2025-07-30

kunal-vaishnavi commented on 2025-07-30

kunal-vaishnavi commented on 2025-07-31

tianleiwu dismissed these changes on 2025-07-31

kunal-vaishnavi dismissed these changes on 2025-07-31

kunal-vaishnavi dismissed their stale review via 6901724e 194 days ago

tianleiwu commented on 2025-07-31

apsonawane force pushed from 8193fcc2 to 0c129dba 193 days ago

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

tianleiwu commented on 2025-08-01

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

Initial QMoE CPU support

3504f148

Fix Lint error

12aa6c39

Fix pipelines

de5e7c55

Add SwiGLU support for CPU QMoE

2f9192ea

Fix pipelines

ddde845e

Address comments

f55d7801

Update contrib ops doc

a1a1f7c3

Update emsdk

337d56a6

Revert "Update emsdk"

0d162528

Address comments

67b4b1fe

Address comments

0fcdc721

Fix

9fdb2ff9

Comments

6e60f098

Update to symmetric quantization

2814dcd0

Fix build errors and update the tests

19588141

Add SwiGLU tests in python

61ab80fd

Add SwiGLU clamping and fix docs pipeline

bd721db4

apsonawane force pushed from d3762dd5 to bd721db4 192 days ago

Add back ep_graph tests

ce1309fa

Revert cuda test changes

0729e443

tianleiwu dismissed these changes on 2025-08-02

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

update doc

271cba4a

tianleiwu dismissed their stale review via 271cba4a 192 days ago

tianleiwu requested a review from

kunal-vaishnavi 192 days ago

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

tianleiwu merged 4004a15b into main 192 days ago

tianleiwu deleted the asonawane/qmoe branch 192 days ago

jywu-msft added release:1.23.0

tianleiwu added cherry-picked

tianleiwu removed release:1.23.0

Reviewers

kunal-vaishnavi

tianleiwu

github-actions

github-advanced-security

Assignees

No one assigned

Labels

cherry-picked

Milestone

No milestone

onnxruntime Add support for QMoE in CPU #25558 Merged

Add support for QMoE in CPU #25558

onnxruntime
Add support for QMoE in CPU
#25558

Merged