Skip to content

[Ulysses] Carry the KV head count per DistributedAttention - #8316

Merged
delock merged 6 commits into
deepspeedai:masterfrom
alanhuangyoo:fix/ulysses-per-model-head-count
Sep 3, 2026
Merged

[Ulysses] Carry the KV head count per DistributedAttention#8316
delock merged 6 commits into
deepspeedai:masterfrom
alanhuangyoo:fix/ulysses-per-model-head-count