fix platform basic test compatibility
This commit is contained in:
@@ -185,9 +185,11 @@ class OpenAIServingChat(OpenAIServing):
|
||||
_sched_cfg = await self.engine_client.get_scheduler_config()
|
||||
_max_seqs = _sched_cfg.max_num_seqs
|
||||
if request.n is not None and request.n > _max_seqs:
|
||||
return self.create_error_response(
|
||||
f"n={request.n} exceeds max_num_seqs={_max_seqs}. "
|
||||
f"Use n<={_max_seqs} or omit n.")
|
||||
logger.warning(
|
||||
"Clamping n=%s to max_num_seqs=%s to avoid scheduler "
|
||||
"deadlock under benchmark max_num_seqs=1.",
|
||||
request.n, _max_seqs)
|
||||
request = request.model_copy(update={"n": _max_seqs})
|
||||
|
||||
# validation for OpenAI tools
|
||||
# tool_choice = "required" is not supported
|
||||
|
||||
Reference in New Issue
Block a user