vllm
[Refactor] Move MXFP8 GEMM management into MxFp8LinearKernel
#39205

Merged

[Refactor] Move MXFP8 GEMM management into MxFp8LinearKernel #39205

vllm-bot merged 5 commits into vllm-project:main from neuralmagic:mxfp8-linear-kernel

[Refactor] Move MXFP8 GEMM management into MxFp8LinearKernel

f9e8cba3

mgoin requested a review from

robertgshaw2-redhat 35 days ago

mgoin requested a review from

tlrmchlsmth 35 days ago

mgoin requested a review from

yewentao256 35 days ago

mgoin requested a review from

pavanimajety 35 days ago

mgoin removed review request from

tlrmchlsmth 35 days ago

mgoin removed review request from

pavanimajety 35 days ago

mgoin removed review request from

yewentao256 35 days ago

mgoin added ready

mgoin added nvidia

mgoin added quantization

gemini-code-assist commented on 2026-04-07

mergify added needs-rebase

Merge origin/main into mxfp8-linear-kernel, resolving conflicts

16245f7e

mergify removed needs-rebase

Merge branch 'main' into mxfp8-linear-kernel

2f7e18ca

mergify added needs-rebase

Merge branch 'main' into mxfp8-linear-kernel

5e4dacbd

Fix Mxfp8OnlineLinearMethod AttributeError on activation_quant_key

e0796502

mergify removed needs-rebase

vllm-bot merged 11e2375f into main 32 days ago

Reviewers

gemini-code-assist

robertgshaw2-redhat

Assignees

No one assigned

Labels

ready nvidia quantization

Milestone

No milestone