EngineX-Hygon/sglang
5
0
Fork 0
Code Issues Pull Requests Actions 7 Projects Releases Wiki Activity

Fix throughput threshold for amd ci test (#6414)

Browse Source
This commit is contained in:
Baizhou Zhang
2025-05-19 14:17:41 -07:00
committed by GitHub
parent 506e5de8fe
commit 299fd22f9e
1 changed files with 5 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
test/srt/test_torch_compile_moe.py
View File

@@ -4,7 +4,7 @@ from types import SimpleNamespace
import requests import requests
from sglang.srt.utils import kill_process_tree from sglang.srt.utils import is_cuda, kill_process_tree
from sglang.test.run_eval import run_eval from sglang.test.run_eval import run_eval
from sglang.test.test_utils import ( from sglang.test.test_utils import (
DEFAULT_SMALL_MOE_MODEL_NAME_FOR_TEST, DEFAULT_SMALL_MOE_MODEL_NAME_FOR_TEST,
@@ -67,7 +67,10 @@ class TestTorchCompileMoe(CustomTestCase):
tok = time.perf_counter() tok = time.perf_counter()
print(f"{res=}") print(f"{res=}")
throughput = max_tokens / (tok - tic) throughput = max_tokens / (tok - tic)
self.assertGreaterEqual(throughput, 285) if is_cuda():
self.assertGreaterEqual(throughput, 285)
else:
self.assertGreaterEqual(throughput, 270)
if __name__ == "__main__": if __name__ == "__main__":