22 lines
653 B
Python
22 lines
653 B
Python
# SPDX-License-Identifier: Apache-2.0
|
|
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
|
|
import argparse
|
|
|
|
from vllm.benchmarks.latency import add_cli_args, main
|
|
from vllm.entrypoints.cli.benchmark.base import BenchmarkSubcommandBase
|
|
|
|
|
|
class BenchmarkLatencySubcommand(BenchmarkSubcommandBase):
|
|
"""The `latency` subcommand for `vllm bench`."""
|
|
|
|
name = "latency"
|
|
help = "Benchmark the latency of a single batch of requests."
|
|
|
|
@classmethod
|
|
def add_cli_args(cls, parser: argparse.ArgumentParser) -> None:
|
|
add_cli_args(parser)
|
|
|
|
@staticmethod
|
|
def cmd(args: argparse.Namespace) -> None:
|
|
main(args)
|