onnxruntime
[CUDA] Bound the FP8 weight dequant scratch by tiling over N
#32129

Merged

[CUDA] Bound the FP8 weight dequant scratch by tiling over N #32129

tianleiwu merged 3 commits into microsoft:main from tianleiwu:tlwu/fp8_dequant_scratch_tiling

[CUDA] Bound the FP8 weight dequant scratch by tiling over N

a3a02167

tianleiwu requested a review from

copilot-pull-request-reviewer 71 days ago

copilot-pull-request-reviewer commented on 2026-08-17

fix(cuda): address FP8 scratch tiling feedback

449dfd9a

test(cuda): enforce FP8 dequant scratch cap

8240f422

tianleiwu requested a review from

baijumeswani 70 days ago

tianleiwu requested a review from

hariharans29 70 days ago

tianleiwu requested a review from

titaiwangms 70 days ago

baijumeswani approved these changes on 2026-08-18

tianleiwu merged 8b9258f1 into main 70 days ago

Reviewers

baijumeswani

hariharans29

titaiwangms

Assignees

No one assigned

Labels

None yet

Milestone

No milestone