[Security] Gate per-request multimodal processor kwargs (#58830)

Signed-off-by: Juan Pérez de Algaba <jperezde@redhat.com>
Co-authored-by: Cursor <cursoragent@cursor.com>
This commit is contained in:
Juan Pérez de Algaba
2026-09-26 11:10:35 +00:00
committed by GitHub
co-authored by Cursor
parent 2f41e00235
commit 31f2e70cd3
17 changed files with 203 additions and 4 deletions
+7
View File
@@ -16,6 +16,7 @@ from vllm import PoolingRequestOutput, envs
from vllm.config import VllmConfig
from vllm.engine.protocol import EngineClient
from vllm.entrypoints.chat_utils import ChatTemplateConfig
from vllm.entrypoints.generate.base.protocol import validate_request_mm_kwargs
from vllm.entrypoints.openai.models.serving import OpenAIServingModels
from vllm.entrypoints.serve.engine.serving import BaseServing
from vllm.entrypoints.serve.engine.typing import AnyRequest
@@ -60,6 +61,7 @@ class PoolingBaseServing(ABC, BaseServing):
self.return_tokens_as_token_ids = return_tokens_as_token_ids
self.log_error_stack = log_error_stack
self.chat_template_config = chat_template_config
self.trust_request_mm_kwargs = chat_template_config.trust_request_mm_kwargs
# Shared thread pool executor for preprocessing and postprocessing.
self._executor: Executor = self.renderer._executor
@@ -108,6 +110,11 @@ class PoolingBaseServing(ABC, BaseServing):
request: AnyPoolingRequest,
raw_request: Request | None = None,
):
validate_request_mm_kwargs(
mm_processor_kwargs=getattr(request, "mm_processor_kwargs", None),
media_io_kwargs=getattr(request, "media_io_kwargs", None),
trust_request_mm_kwargs=self.trust_request_mm_kwargs,
)
base_request_id = self._base_request_id(
raw_request, getattr(request, "request_id", None)
)
+1
View File
@@ -166,6 +166,7 @@ def init_pooling_state(
chat_template=resolved_chat_template,
chat_template_content_format=args.chat_template_content_format,
trust_request_chat_template=args.trust_request_chat_template,
trust_request_mm_kwargs=args.trust_request_mm_kwargs,
)
state.serving_pooling = (