From 459764104d2f8ac1b8a947f9a5a2603bdb07e360 Mon Sep 17 00:00:00 2001 From: "wang.yuqi" Date: Thu, 8 Oct 2026 17:56:05 +0800 Subject: [PATCH] [Frontend] Move offline_utils.py to entrypoints/common. (#58052) Signed-off-by: wang.yuqi --- .github/CODEOWNERS | 2 +- tests/entrypoints/multimodal/llm/test_mm_processor_kwargs.py | 4 ++-- tests/entrypoints/unit_tests/test_offline_utils.py | 2 +- vllm/entrypoints/common/__init__.py | 0 vllm/entrypoints/{offline_utils.py => common/offline.py} | 0 vllm/entrypoints/generate/beam_search/offline.py | 2 +- vllm/entrypoints/llm.py | 2 +- vllm/entrypoints/pooling/offline.py | 2 +- vllm/entrypoints/rl/offline.py | 2 +- 9 files changed, 8 insertions(+), 8 deletions(-) create mode 100644 vllm/entrypoints/common/__init__.py rename vllm/entrypoints/{offline_utils.py => common/offline.py} (100%) diff --git a/.github/CODEOWNERS b/.github/CODEOWNERS index 677e37ff8b..65d9454b28 100644 --- a/.github/CODEOWNERS +++ b/.github/CODEOWNERS @@ -45,7 +45,7 @@ /vllm/entrypoints/serve @njhill /vllm/entrypoints/*.py @njhill /vllm/entrypoints/chat_utils.py @DarkLight1337 -/vllm/entrypoints/offline_utils.py @DarkLight1337 +/vllm/entrypoints/common @DarkLight1337 /vllm/entrypoints/llm.py @DarkLight1337 # Rust Frontend diff --git a/tests/entrypoints/multimodal/llm/test_mm_processor_kwargs.py b/tests/entrypoints/multimodal/llm/test_mm_processor_kwargs.py index c03efa8f10..55990ed1dd 100644 --- a/tests/entrypoints/multimodal/llm/test_mm_processor_kwargs.py +++ b/tests/entrypoints/multimodal/llm/test_mm_processor_kwargs.py @@ -212,7 +212,7 @@ def test_preprocess_cmpl_applies_mm_processor_kwargs_to_renderer( llm.renderer = renderer monkeypatch.setattr( - "vllm.entrypoints.offline_utils.parse_model_prompt", + "vllm.entrypoints.common.offline.parse_model_prompt", lambda _model_config, parsed_prompt: parsed_prompt, ) @@ -246,7 +246,7 @@ def test_preprocess_cmpl_keeps_prompt_mm_processor_kwargs_when_no_override( llm.renderer = renderer monkeypatch.setattr( - "vllm.entrypoints.offline_utils.parse_model_prompt", + "vllm.entrypoints.common.offline.parse_model_prompt", lambda _model_config, parsed_prompt: parsed_prompt, ) diff --git a/tests/entrypoints/unit_tests/test_offline_utils.py b/tests/entrypoints/unit_tests/test_offline_utils.py index 39f77672a4..a37969c9b3 100644 --- a/tests/entrypoints/unit_tests/test_offline_utils.py +++ b/tests/entrypoints/unit_tests/test_offline_utils.py @@ -4,7 +4,7 @@ import pytest from vllm import SamplingParams -from vllm.entrypoints.offline_utils import OfflineInferenceMixin +from vllm.entrypoints.common.offline import OfflineInferenceMixin from vllm.exceptions import VLLMValidationError from vllm.lora.request import LoRARequest diff --git a/vllm/entrypoints/common/__init__.py b/vllm/entrypoints/common/__init__.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/vllm/entrypoints/offline_utils.py b/vllm/entrypoints/common/offline.py similarity index 100% rename from vllm/entrypoints/offline_utils.py rename to vllm/entrypoints/common/offline.py diff --git a/vllm/entrypoints/generate/beam_search/offline.py b/vllm/entrypoints/generate/beam_search/offline.py index 1ad451ac13..0acd0b43a5 100644 --- a/vllm/entrypoints/generate/beam_search/offline.py +++ b/vllm/entrypoints/generate/beam_search/offline.py @@ -8,7 +8,7 @@ import torch from tqdm import tqdm from vllm import RequestOutput, TextPrompt, TokensPrompt -from vllm.entrypoints.offline_utils import OfflineInferenceMixin +from vllm.entrypoints.common.offline import OfflineInferenceMixin from vllm.logger import init_logger from vllm.lora.request import LoRARequest from vllm.pooling_params import PoolingParams diff --git a/vllm/entrypoints/llm.py b/vllm/entrypoints/llm.py index 9dfff78ed9..a879b61484 100644 --- a/vllm/entrypoints/llm.py +++ b/vllm/entrypoints/llm.py @@ -33,6 +33,7 @@ from vllm.entrypoints.chat_utils import ( ChatTemplateContentFormatOption, load_chat_template, ) +from vllm.entrypoints.common.offline import _O, _R, OfflineInferenceMixin from vllm.entrypoints.generate.beam_search.offline import BeamSearchOfflineMixin from vllm.entrypoints.pooling.offline import PoolingOfflineMixin from vllm.entrypoints.rl.offline import RLOfflineMixin @@ -52,7 +53,6 @@ from vllm.v1.engine.llm_engine import LLMEngine from vllm.v1.sample.logits_processor import LogitsProcessor from ..renderers import ChatParams -from .offline_utils import _O, _R, OfflineInferenceMixin if TYPE_CHECKING: from vllm.v1.metrics.reader import Metric diff --git a/vllm/entrypoints/pooling/offline.py b/vllm/entrypoints/pooling/offline.py index 27388f1ff0..81d78101c4 100644 --- a/vllm/entrypoints/pooling/offline.py +++ b/vllm/entrypoints/pooling/offline.py @@ -7,7 +7,7 @@ from typing import Any from tqdm.auto import tqdm from vllm.entrypoints.chat_utils import ChatTemplateConfig -from vllm.entrypoints.offline_utils import OfflineInferenceMixin +from vllm.entrypoints.common.offline import OfflineInferenceMixin from vllm.inputs import DataPrompt, PromptType from vllm.logger import init_logger from vllm.lora.request import LoRARequest diff --git a/vllm/entrypoints/rl/offline.py b/vllm/entrypoints/rl/offline.py index 02fc63104d..1db6a82533 100644 --- a/vllm/entrypoints/rl/offline.py +++ b/vllm/entrypoints/rl/offline.py @@ -5,7 +5,7 @@ from vllm.distributed.weight_transfer.base import ( WeightTransferInitRequest, WeightTransferUpdateRequest, ) -from vllm.entrypoints.offline_utils import OfflineInferenceMixin +from vllm.entrypoints.common.offline import OfflineInferenceMixin class RLOfflineMixin(OfflineInferenceMixin):