Rename sglang.bench_latency to sglang.bench_one_batch (#2118)

This commit is contained in:
Lianmin Zheng
2024-11-21 20:07:48 -08:00
committed by GitHub
parent 8048c28c11
commit dfec7fca06
16 changed files with 521 additions and 599 deletions

View File

@@ -0,0 +1,28 @@
import unittest
from sglang.test.test_utils import (
DEFAULT_MODEL_NAME_FOR_TEST,
DEFAULT_MOE_MODEL_NAME_FOR_TEST,
is_in_ci,
run_bench_one_batch,
)
class TestBenchOneBatch(unittest.TestCase):
def test_default(self):
output_throughput = run_bench_one_batch(DEFAULT_MODEL_NAME_FOR_TEST, [])
if is_in_ci():
self.assertGreater(output_throughput, 135)
def test_moe_default(self):
output_throughput = run_bench_one_batch(
DEFAULT_MOE_MODEL_NAME_FOR_TEST, ["--tp", "2"]
)
if is_in_ci():
self.assertGreater(output_throughput, 125)
if __name__ == "__main__":
unittest.main()