[Doc] Add docs about OpenAI compatible server (#3288)

vllm/entrypoints/openai/api_server.py

@@ -1,11 +1,8 @@
-import argparse
 import asyncio
-import json
 from contextlib import asynccontextmanager
 import os
 import importlib
 import inspect
-import ssl
 
 from prometheus_client import make_asgi_app
 import fastapi
@@ -23,9 +20,9 @@ from vllm.entrypoints.openai.protocol import (CompletionRequest,
                                               ChatCompletionRequest,
                                               ErrorResponse)
 from vllm.logger import init_logger
+from vllm.entrypoints.openai.cli_args import make_arg_parser
 from vllm.entrypoints.openai.serving_chat import OpenAIServingChat
 from vllm.entrypoints.openai.serving_completion import OpenAIServingCompletion
-from vllm.entrypoints.openai.serving_engine import LoRA
 
 TIMEOUT_KEEP_ALIVE = 5  # seconds
 
@@ -51,109 +48,8 @@ async def lifespan(app: fastapi.FastAPI):
 app = fastapi.FastAPI(lifespan=lifespan)
 
 
-class LoRAParserAction(argparse.Action):
-
-    def __call__(self, parser, namespace, values, option_string=None):
-        lora_list = []
-        for item in values:
-            name, path = item.split('=')
-            lora_list.append(LoRA(name, path))
-        setattr(namespace, self.dest, lora_list)
-
-
 def parse_args():
-    parser = argparse.ArgumentParser(
-        description="vLLM OpenAI-Compatible RESTful API server.")
-    parser.add_argument("--host", type=str, default=None, help="host name")
-    parser.add_argument("--port", type=int, default=8000, help="port number")
-    parser.add_argument(
-        "--uvicorn-log-level",
-        type=str,
-        default="info",
-        choices=['debug', 'info', 'warning', 'error', 'critical', 'trace'],
-        help="log level for uvicorn")
-    parser.add_argument("--allow-credentials",
-                        action="store_true",
-                        help="allow credentials")
-    parser.add_argument("--allowed-origins",
-                        type=json.loads,
-                        default=["*"],
-                        help="allowed origins")
-    parser.add_argument("--allowed-methods",
-                        type=json.loads,
-                        default=["*"],
-                        help="allowed methods")
-    parser.add_argument("--allowed-headers",
-                        type=json.loads,
-                        default=["*"],
-                        help="allowed headers")
-    parser.add_argument("--api-key",
-                        type=str,
-                        default=None,
-                        help="If provided, the server will require this key "
-                        "to be presented in the header.")
-    parser.add_argument("--served-model-name",
-                        type=str,
-                        default=None,
-                        help="The model name used in the API. If not "
-                        "specified, the model name will be the same as "
-                        "the huggingface name.")
-    parser.add_argument(
-        "--lora-modules",
-        type=str,
-        default=None,
-        nargs='+',
-        action=LoRAParserAction,
-        help="LoRA module configurations in the format name=path. "
-        "Multiple modules can be specified.")
-    parser.add_argument("--chat-template",
-                        type=str,
-                        default=None,
-                        help="The file path to the chat template, "
-                        "or the template in single-line form "
-                        "for the specified model")
-    parser.add_argument("--response-role",
-                        type=str,
-                        default="assistant",
-                        help="The role name to return if "
-                        "`request.add_generation_prompt=true`.")
-    parser.add_argument("--ssl-keyfile",
-                        type=str,
-                        default=None,
-                        help="The file path to the SSL key file")
-    parser.add_argument("--ssl-certfile",
-                        type=str,
-                        default=None,
-                        help="The file path to the SSL cert file")
-    parser.add_argument("--ssl-ca-certs",
-                        type=str,
-                        default=None,
-                        help="The CA certificates file")
-    parser.add_argument(
-        "--ssl-cert-reqs",
-        type=int,
-        default=int(ssl.CERT_NONE),
-        help="Whether client certificate is required (see stdlib ssl module's)"
-    )
-    parser.add_argument(
-        "--root-path",
-        type=str,
-        default=None,
-        help="FastAPI root_path when app is behind a path based routing proxy")
-    parser.add_argument(
-        "--middleware",
-        type=str,
-        action="append",
-        default=[],
-        help="Additional ASGI middleware to apply to the app. "
-        "We accept multiple --middleware arguments. "
-        "The value should be an import path. "
-        "If a function is provided, vLLM will add it to the server "
-        "using @app.middleware('http'). "
-        "If a class is provided, vLLM will add it to the server "
-        "using app.add_middleware(). ")
-
-    parser = AsyncEngineArgs.add_cli_args(parser)
+    parser = make_arg_parser()
     return parser.parse_args()