PR #18653 CANN: support gated linear attn

CANN: support gated linear attn #18653

hipudding merged 3 commits into ggml-org:master from hipudding:gla

hipudding marked this pull request as draft 156 days ago

github-actions added ggml

github-actions added Ascend NPU

hipudding force pushed from 7c21f715 to 81f18f3c 156 days ago

hipudding force pushed from 81f18f3c to 746a693d 156 days ago

hipudding marked this pull request as ready for review 156 days ago

hipudding requested a review from

noemotiovon 156 days ago

noemotiovon approved these changes on 2026-01-07

hipudding requested a review from

noemotiovon 150 days ago

hipudding requested a review from

ggerganov 150 days ago

noemotiovon approved these changes on 2026-01-13

CANN: support gated linear attn

4f81fc99

CANN: optimize OP gla

0809aa59

Remove unused comments

bfa67a8a

hipudding force pushed from 903d0ce8 to bfa67a8a 147 days ago

ggerganov approved these changes on 2026-01-16

hipudding merged baa4ba0a into master 147 days ago

Reviewers

ggerganov

noemotiovon

Assignees

No one assigned

Labels

ggml Ascend NPU

Milestone

No milestone