Bugfix: LMCache Connector with Sglang (#12946)

Signed-off-by: Mohd Muzzammil <me.muzzammil@samsung.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
This commit is contained in:
MMuzzammil1
2025-11-10 13:13:37 +09:00
committed by GitHub
parent 61c7fe7aed
commit 1f2a6c691b

View File

@@ -107,7 +107,7 @@ class LMCRadixCache(RadixCache):
"v_buffer",
getattr(self.token_to_kv_pool_allocator._kvcache, "v_buffer"),
),
tp_group=tp_group,
tp_group=tp_group.device_group if tp_group is not None else None,
)
self.load_stream = torch.cuda.Stream()