onnxruntime
WIP: DP4AMatMul fix matmul for subgoup size 64 GPUs
#23637
Merged

WIP: DP4AMatMul fix matmul for subgoup size 64 GPUs #23637

sushraja-msft
sushraja-msft sushraja-msft requested a review from guschmue guschmue 1 year ago
sushraja-msft
qjia7
qjia7 commented on 2025-02-11
guschmue guschmue added ep:WebGPU
sushraja-msft sushraja-msft changed the title DP4AMatMul fix matmul for subgoup size 64 GPUs WIP: DP4AMatMul fix matmul for subgoup size 64 GPUs 1 year ago
sushraja-msft sushraja-msft force pushed from e04ddb2b to ecbabd80 1 year ago
sushraja-msft sushraja-msft force pushed from 1ab5a6a3 to 1decc489 1 year ago
sushraja-msft Do not use subgroups for quantization
4f473cb4
sushraja-msft sushraja-msft force pushed from 1decc489 to 4f473cb4 1 year ago
qjia7
qjia7 commented on 2025-02-13
guschmue
guschmue approved these changes on 2025-02-13
guschmue guschmue merged 4e24d372 into main 1 year ago
guschmue guschmue deleted the user/sushraja/fix_dp4_quantization branch 1 year ago

Login to write a write a comment.

Login via GitHub

Reviewers
Assignees
No one assigned
Labels
Milestone