llama.cpp
hexagon: KV-head team split for multicore HMX flash attention
#31055

Merged

hexagon: KV-head team split for multicore HMX flash attention #31055

max-krasnyansky merged 2 commits into ggml-org:master from qualcomm:ebateni/adaptive-split

ebateni requested a review from

ggerganov 5 days ago

ebateni requested a review 5 days ago

github-actions added testing

github-actions added ggml

github-actions added Hexagon

ebateni force pushed from b73c2a4c to ba02704d 5 days ago

hexagon: KV-head team split for HMX flash attention in multicore

df795911

hexagon: merge FA_HEAD_SPLIT and FA_SPLIT into single GGML_HEXAGON_FA…

33ee7351

max-krasnyansky force pushed from ba02704d to 33ee7351 4 days ago

max-krasnyansky approved these changes on 2026-10-24

max-krasnyansky added merge ready

lhez approved these changes on 2026-10-24

max-krasnyansky merged 78a8ae14 into master 4 days ago

Reviewers

lhez

max-krasnyansky

ggerganov

Assignees

No one assigned

Labels

testing ggml merge ready Hexagon

Milestone

No milestone