Add vllm bench [latency, throughput] CLI commands (#16508)

Signed-off-by: mgoin <mgoin64@gmail.com>
This commit is contained in:
Michael Goin
2025-04-14 23:10:35 -07:00
committed by GitHub
parent bc5dd4f669
commit b4fe16c75b
11 changed files with 1771 additions and 2 deletions
+4 -2
View File
@@ -1,14 +1,16 @@
# SPDX-License-Identifier: Apache-2.0
import argparse
import vllm.entrypoints.cli.benchmark.latency
import vllm.entrypoints.cli.benchmark.serve
import vllm.entrypoints.cli.benchmark.throughput
from vllm.entrypoints.cli.types import CLISubcommand
from vllm.utils import FlexibleArgumentParser
# TODO: Add the rest of the benchmark subcommands here,
# e.g., throughput, latency, etc.
BENCHMARK_CMD_MODULES = [
vllm.entrypoints.cli.benchmark.latency,
vllm.entrypoints.cli.benchmark.serve,
vllm.entrypoints.cli.benchmark.throughput,
]