c781db0f6c
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com> Co-authored-by: cy <chenyang08056032@163.com>
40 lines
1.0 KiB
Python
40 lines
1.0 KiB
Python
import unittest
|
|
|
|
from sglang.test.ascend.gsm8k_ascend_mixin import GSM8KAscendMixin
|
|
from sglang.test.ascend.test_ascend_utils import (
|
|
Qwen3_30B_A3B_Instruct_2507_WEIGHTS_PATH,
|
|
)
|
|
from sglang.test.ci.ci_register import register_npu_ci
|
|
from sglang.test.test_utils import CustomTestCase
|
|
|
|
register_npu_ci(est_time=400, suite="nightly-2-npu-a3", nightly=True)
|
|
|
|
|
|
class TestQwen330B(GSM8KAscendMixin, CustomTestCase):
|
|
"""Testcase: Verify that the inference accuracy of the Qwen/Qwen3-30B-A3B-Instruct-2507 model on the GSM8K dataset is no less than 0.90.
|
|
|
|
[Test Category] Model
|
|
[Test Target] Qwen/Qwen3-30B-A3B-Instruct-2507
|
|
"""
|
|
|
|
model = Qwen3_30B_A3B_Instruct_2507_WEIGHTS_PATH
|
|
accuracy = 0.90
|
|
other_args = [
|
|
"--trust-remote-code",
|
|
"--mem-fraction-static",
|
|
0.7,
|
|
"--max-running-requests",
|
|
32,
|
|
"--attention-backend",
|
|
"ascend",
|
|
"--disable-cuda-graph",
|
|
"--cuda-graph-max-bs",
|
|
32,
|
|
"--tp-size",
|
|
2,
|
|
]
|
|
|
|
|
|
if __name__ == "__main__":
|
|
unittest.main()
|