[Frontend] Cleanup serving engine (#33103)

Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
This commit is contained in:
Cyrus Leung
2026-01-26 20:47:26 -08:00
committed by GitHub
parent 3b8f0fe59e
commit e0b005d9cf
5 changed files with 57 additions and 40 deletions
+6 -2
View File
@@ -116,6 +116,7 @@ from vllm.entrypoints.openai.responses.utils import (
extract_tool_types,
should_continue_final_message,
)
from vllm.entrypoints.utils import get_max_tokens
from vllm.exceptions import VLLMValidationError
from vllm.inputs.data import TokensPrompt
from vllm.logger import init_logger
@@ -435,8 +436,11 @@ class OpenAIServingResponses(OpenAIServing):
if maybe_error is not None:
return maybe_error
default_max_tokens = self.max_model_len - len(
engine_prompt["prompt_token_ids"]
default_max_tokens = get_max_tokens(
self.max_model_len,
request,
engine_prompt,
self.default_sampling_params,
)
sampling_params = request.to_sampling_params(