chenchenghao/sglang
1
0
Fork 0
Code Issues Pull Requests Actions 12 Packages Projects Releases Wiki Activity

ci: speed up b200 ci (#13237)

Browse Source
This commit is contained in:
b8zhong
2025-11-13 23:57:33 -08:00
committed by GitHub
parent 385ff0e56f
commit 7b79cc4fe2
1 changed files with 6 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
test/srt/test_deepseek_v3_fp4_4gpu.py
View File

@@ -35,6 +35,8 @@ class TestDeepseekV3FP4(CustomTestCase):
"modelopt_fp4",
"--kv-cache-dtype",
"fp8_e4m3",
"--model-loader-extra-config",
'{"enable_multithread_load": true}',
]
cls.process = popen_launch_server(
cls.model,
@@ -106,6 +108,8 @@ class TestDeepseekV3FP4MTP(CustomTestCase):
"4",
"--kv-cache-dtype",
"fp8_e4m3",
"--model-loader-extra-config",
'{"enable_multithread_load": true}',
]
cls.process = popen_launch_server(
cls.model,
@@ -184,6 +188,8 @@ class TestDeepseekV3FP4CutlassMoE(CustomTestCase):
"flashinfer_cutlass",
"--quantization",
"modelopt_fp4",
"--model-loader-extra-config",
'{"enable_multithread_load": true}',
]
cls.process = popen_launch_server(
cls.model,