vllm.models.glm5next ¶
Modules:
-
amd– -
common– -
nvidia– -
sparse_indexer–Sparse attention indexer (kpool) — hardware-isolated entry point.
vllm.models.glm5next ¶Modules:
amd – common – nvidia – sparse_indexer – Sparse attention indexer (kpool) — hardware-isolated entry point.