INNER CODE UNIT · Python
run_server
raullenchai/Rapid-MLX · vllm_mlx/models/deepseek_v41_native/server.py:26
def run_server(
*,
host: str,
port: int,
served_model_name: str,
default_max_tokens: int,
cors_origins: list[str],
uvicorn_log_level: str,
no_thinking: bool = False,
api_key: str | None = None,
rate_limit: int = 0,
max_request_bytes: int = 8 * 1024 * 1024,
body_receive_timeout_seconds: float = 15.0,
default_timeout: float = 1800.0,
max_concurrent_requests: int = 256,
cors_policy: Any | None = None,
reasoning_parser_name: str | None = "deepseek_v4",
) -> None: