onnxruntime
[CUDA] Support SwiGlu in MoE and qMoE
#25530
Merged

[CUDA] Support SwiGlu in MoE and qMoE #25530

tianleiwu merged 6 commits into main from tlwu/moe_swiglu
tianleiwu
tianleiwu support swiglu
755e33c7
tianleiwu only build interleaved version
514b072d
tianleiwu Update qMoE
63d8808d
tianleiwu tianleiwu changed the title [CUDA] Support SwiGlu in MoE [CUDA] Support SwiGlu in MoE and qMoE 226 days ago
tianleiwu Update operator doc
e24271e8
tianleiwu static cast
4c07bcf6
tianleiwu add test for 4 bits
360f4d0e
apsonawane
apsonawane approved these changes on 2025-07-28
tianleiwu tianleiwu merged a2b4546c into main 225 days ago
tianleiwu tianleiwu deleted the tlwu/moe_swiglu branch 225 days ago
jywu-msft jywu-msft added release:1.23.0
tianleiwu tianleiwu added cherry-picked
tianleiwu tianleiwu removed release:1.23.0

Login to write a write a comment.

Login via GitHub

Reviewers
Assignees
No one assigned
Labels
Milestone