biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[V0 Deprecation] Remove LLMEngine (#25033)

Browse Source 
Signed-off-by: Woosuk Kwon <woosuk@thinkingmachines.ai>
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
This commit is contained in:
Woosuk Kwon
2025-09-20 17:56:30 -07:00
committed by GitHub
parent 367a480bd3
commit 52c2a8d4ad
29 changed files with 65 additions and 2763 deletions
Download Patch File Download Diff File Expand all files Collapse all files

22
vllm/model_executor/model_loader/tensorizer.py
View File

@@ -672,21 +672,15 @@ def tensorize_vllm_model(engine_args: "EngineArgs",
) as stream:
stream.write(encryption_params.key)
from vllm import LLMEngine
from vllm.v1.engine.llm_engine import LLMEngine as V1LLMEngine
assert envs.VLLM_USE_V1
if not envs.VLLM_USE_V1:
engine = LLMEngine.from_engine_args(engine_args)
engine.model_executor.collective_rpc(
"save_tensorized_model",
kwargs={"tensorizer_config": tensorizer_config.to_serializable()},
)
else:
engine = V1LLMEngine.from_vllm_config(engine_config)
engine.collective_rpc(
"save_tensorized_model",
kwargs={"tensorizer_config": tensorizer_config.to_serializable()},
)
from vllm.v1.engine.llm_engine import LLMEngine
engine = LLMEngine.from_vllm_config(engine_config)
engine.collective_rpc(
"save_tensorized_model",
kwargs={"tensorizer_config": tensorizer_config.to_serializable()},
)
def tensorize_lora_adapter(lora_path: str,