Files
vllm/tests/benchmarks/test_throughput_cli.py
2025-10-05 07:06:22 -07:00

31 lines
694 B
Python

# SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import subprocess
import pytest
MODEL_NAME = "meta-llama/Llama-3.2-1B-Instruct"
@pytest.mark.benchmark
def test_bench_throughput():
command = [
"vllm",
"bench",
"throughput",
"--model",
MODEL_NAME,
"--input-len",
"32",
"--output-len",
"1",
"--enforce-eager",
"--load-format",
"dummy",
]
result = subprocess.run(command, capture_output=True, text=True)
print(result.stdout)
print(result.stderr)
assert result.returncode == 0, f"Benchmark failed: {result.stderr}"