[Frontend] Make finish_reason and Responses item status truthful (#59324)

Signed-off-by: Flora Feng <4florafeng@gmail.com>
Signed-off-by: sfeng33 <4florafeng@gmail.com>
Co-authored-by: Chauncey <chaunceyjiang@gmail.com>
This commit is contained in:
Flora Feng
2026-10-15 10:12:02 +01:00
committed by GitHub
co-authored by Chauncey
parent 427915bed2
commit 80dd705c1d
14 changed files with 421 additions and 62 deletions
+3 -1
View File
@@ -32,6 +32,7 @@ from vllm.entrypoints.openai.responses.protocol import (
from vllm.entrypoints.openai.responses.utils import (
build_response_output_items,
construct_tool_dicts,
response_item_status,
)
from vllm.entrypoints.serve.utils.tool_calls_utils import (
maybe_filter_parallel_tool_calls,
@@ -380,6 +381,7 @@ class ParsableContext(ConversationContext):
tool_calls or [], self.request
),
tools=self.request.tools,
finish_reason=completion.finish_reason,
)
)
elif completion.text:
@@ -387,7 +389,7 @@ class ParsableContext(ConversationContext):
ResponseOutputMessage(
type="message",
id=f"msg_{random_uuid()}",
status="completed",
status=response_item_status(completion.finish_reason),
role="assistant",
content=[
ResponseOutputText(