ggml-webgpu: makes the flash attn vec path subgroup-aware #23040
ggml-webgpu: makes the flash attn vec path compile and size its split…
4c0f6291
ggml-webgpu: remove the extra max_wg_size >= max_subgroup_size guard.…
d53c0c1b
CISC
approved these changes
on 2026-05-14
Assignees
No one assigned
Login to write a write a comment.
Login via GitHub