DeepSpeed
Add fused triton kernel for swiglu
#8244
Merged

Add fused triton kernel for swiglu #8244

pengdurice merged 6 commits into master from hongwei/swiglu
hwchen2017
hwchen2017 hwchen2017 requested a review from delock delock 16 days ago
hwchen2017 hwchen2017 requested a review from tohtana tohtana 16 days ago
hwchen2017 hwchen2017 requested a review from sfc-gh-truwase sfc-gh-truwase 16 days ago
hwchen2017 hwchen2017 requested a review from tjruwase tjruwase 16 days ago
hwchen2017 hwchen2017 requested a review from loadams loadams 16 days ago
chatgpt-codex-connector
chatgpt-codex-connector commented on 2026-08-11
sfc-gh-truwase
sfc-gh-truwase commented on 2026-08-11
pengdurice
pengdurice commented on 2026-08-12
pengdurice
pengdurice commented on 2026-08-12
pengdurice
hwchen2017 add triton swiglu kernel; move triton ops to oen folder
b1a5b390
hwchen2017 resue triton availability check
2404ed2b
hwchen2017 Add benchmark code
da3413f0
hwchen2017 Add sanity check
3b527f1b
hwchen2017 Fix reviewer's comments
1df56d2b
hwchen2017 Fix CI error by using smaller range for input
7fc5b57c
hwchen2017 hwchen2017 force pushed from d7630fed to 7fc5b57c 6 days ago
hwchen2017
pengdurice
pengdurice approved these changes on 2026-08-21
pengdurice pengdurice merged 24a41afa into master 6 days ago
pengdurice pengdurice deleted the hongwei/swiglu branch 6 days ago

Login to write a write a comment.

Login via GitHub

Assignees
No one assigned
Labels
Milestone