Add full serve CLI reference back to docs (#20978)

Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
This commit is contained in:
Harry Mellor
2025-07-15 17:42:30 +00:00
committed by GitHub
parent 1e36c8687e
commit b637e9dcb8
6 changed files with 58 additions and 35 deletions
-31
View File
@@ -67,37 +67,6 @@ class ServeSubcommand(CLISubcommand):
help="Start the vLLM OpenAI Compatible API server.",
description="Start the vLLM OpenAI Compatible API server.",
usage="vllm serve [model_tag] [options]")
serve_parser.add_argument("model_tag",
type=str,
nargs='?',
help="The model tag to serve "
"(optional if specified in config)")
serve_parser.add_argument(
"--headless",
action='store_true',
default=False,
help="Run in headless mode. See multi-node data parallel "
"documentation for more details.")
serve_parser.add_argument(
'--data-parallel-start-rank',
'-dpr',
type=int,
default=0,
help="Starting data parallel rank for secondary nodes. "
"Requires --headless.")
serve_parser.add_argument('--api-server-count',
'-asc',
type=int,
default=1,
help='How many API server processes to run.')
serve_parser.add_argument(
"--config",
type=str,
default='',
required=False,
help="Read CLI options from a config file. "
"Must be a YAML with the following options: "
"https://docs.vllm.ai/en/latest/configuration/serve_args.html")
serve_parser = make_arg_parser(serve_parser)
show_filtered_argument_or_group_from_help(serve_parser, ["serve"])
+28
View File
@@ -236,6 +236,34 @@ def make_arg_parser(parser: FlexibleArgumentParser) -> FlexibleArgumentParser:
register all arguments instead of manually enumerating them here. This
avoids code duplication and keeps the argument definitions in one place.
"""
parser.add_argument("model_tag",
type=str,
nargs="?",
help="The model tag to serve "
"(optional if specified in config)")
parser.add_argument(
"--headless",
action="store_true",
default=False,
help="Run in headless mode. See multi-node data parallel "
"documentation for more details.")
parser.add_argument(
"--data-parallel-start-rank",
"-dpr",
type=int,
default=0,
help="Starting data parallel rank for secondary nodes. "
"Requires --headless.")
parser.add_argument("--api-server-count",
"-asc",
type=int,
default=1,
help="How many API server processes to run.")
parser.add_argument(
"--config",
help="Read CLI options from a config file. "
"Must be a YAML with the following options: "
"https://docs.vllm.ai/en/latest/configuration/serve_args.html")
parser = FrontendArgs.add_cli_args(parser)
parser = AsyncEngineArgs.add_cli_args(parser)