support non disturbing remote instance weight loader v2 (#14997)

Signed-off-by: Anqi Shen <amy.saq@antgroup.com>
This commit is contained in:
amysaq2023
2025-12-16 14:39:56 -08:00
committed by GitHub
parent a4c762811a
commit ccc8f3b266
11 changed files with 557 additions and 40 deletions
+6
View File
@@ -366,6 +366,12 @@ class TpModelWorker(BaseTpWorker):
can_run_cuda_graph=can_run_cuda_graph,
)
def get_remote_instance_transfer_engine_info(self):
return (
self.model_runner.remote_instance_transfer_engine_session_id,
self.model_runner.remote_instance_transfer_engine_weight_info,
)
def forward_batch_generation(
self,
model_worker_batch: ModelWorkerBatch,