[Frontend] Respect Chat Completion parallel_tool_calls param (#26233)

Signed-off-by: Ben Browning <bbrownin@redhat.com>
Co-authored-by: Chauncey <chaunceyjiang@gmail.com>
This commit is contained in:
Ben Browning
2025-11-25 09:56:15 +00:00
committed by GitHub
co-authored by Chauncey
parent a685b47c57
commit e1dd706cd1
6 changed files with 102 additions and 8 deletions
+3
View File
@@ -55,6 +55,7 @@ from vllm.entrypoints.openai.serving_engine import OpenAIServing, clamp_prompt_l
from vllm.entrypoints.openai.serving_models import OpenAIServingModels
from vllm.entrypoints.openai.tool_parsers import ToolParser
from vllm.entrypoints.openai.tool_parsers.mistral_tool_parser import MistralToolCall
from vllm.entrypoints.openai.utils import maybe_filter_parallel_tool_calls
from vllm.entrypoints.utils import get_max_tokens, should_include_usage
from vllm.inputs.data import TokensPrompt as EngineTokensPrompt
from vllm.logger import init_logger
@@ -1206,6 +1207,7 @@ class OpenAIServingChat(OpenAIServing):
finish_reason_sent[i] = True
choice_data = maybe_filter_parallel_tool_calls(choice_data, request)
chunk = ChatCompletionStreamResponse(
id=request_id,
object=chunk_object_type,
@@ -1531,6 +1533,7 @@ class OpenAIServingChat(OpenAIServing):
as_list(output.token_ids) if request.return_token_ids else None
),
)
choice_data = maybe_filter_parallel_tool_calls(choice_data, request)
choices.append(choice_data)