[MRV2] Support stock torch.compile mode (#59079)
Signed-off-by: cheesecake <farzanaman99@gmail.com> Signed-off-by: Nick Hill <nickhill123@gmail.com> Co-authored-by: OpenAI Codex <codex@openai.com> Co-authored-by: Nick Hill <nickhill123@gmail.com> Co-authored-by: Wentao Ye <44945378+yewentao256@users.noreply.github.com>
This commit is contained in:
co-authored by
OpenAI Codex
Nick Hill
Wentao Ye
parent
4406874e7c
commit
4e59d53ab5
@@ -7,6 +7,7 @@ from typing import Any
|
||||
|
||||
import torch
|
||||
|
||||
from vllm.config import CompilationConfig, CompilationMode
|
||||
from vllm.model_executor.warmup.jit_warmup import JitWarmupRegistry
|
||||
from vllm.v1.outputs import EMPTY_MODEL_RUNNER_OUTPUT
|
||||
from vllm.v1.worker.gpu import eplb_utils as eplb
|
||||
@@ -68,7 +69,9 @@ def _make_runner(**overrides: Any) -> Any:
|
||||
enable_batch_sharded_sampling=False,
|
||||
eplb_config=SimpleNamespace(log_balancedness=True),
|
||||
)
|
||||
runner.compilation_config = CompilationConfig(mode=CompilationMode.NONE)
|
||||
runner.vllm_config = SimpleNamespace(
|
||||
compilation_config=runner.compilation_config,
|
||||
load_config=runner.load_config,
|
||||
model_config=runner.model_config,
|
||||
)
|
||||
|
||||
Reference in New Issue
Block a user