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

[CI] Fix mxfp8 TrtllmGenMoe test (#23125)

Browse Source 
Co-authored-by: Brayden Zhong <b8zhong@uwaterloo.ca>
This commit is contained in:
Ziang Li
2026-04-24 09:02:11 +00:00
committed by GitHub
co-authored by Brayden Zhong
parent 92bb5c6bbe
commit 1758856762
1 changed files with 2 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files
test/registered/backends/test_flashinfer_trtllm_gen_moe_backend.py
+2 -4
View File
@@ -115,7 +115,7 @@ class FlashinferTrtllmGenMoeBackendMXFP8Base:
@classmethod @classmethod
def setUpClass(cls): def setUpClass(cls):
cls.model = "Qwen/Qwen3-30B-A3B-Instruct-2507" cls.model = "zianglih/Qwen3-30B-A3B-Instruct-2507-MXFP8"
cls.base_url = DEFAULT_URL_FOR_TEST cls.base_url = DEFAULT_URL_FOR_TEST
cls.process = popen_launch_server( cls.process = popen_launch_server(
cls.model, cls.model,
@@ -123,10 +123,8 @@ class FlashinferTrtllmGenMoeBackendMXFP8Base:
timeout=DEFAULT_TIMEOUT_FOR_SERVER_LAUNCH, timeout=DEFAULT_TIMEOUT_FOR_SERVER_LAUNCH,
env={**os.environ, "SGLANG_ENABLE_JIT_DEEPGEMM": "False"}, env={**os.environ, "SGLANG_ENABLE_JIT_DEEPGEMM": "False"},
other_args=[ other_args=[
"--quantization",
"mxfp8",
"--fp8-gemm-backend", "--fp8-gemm-backend",
"flashinfer_trtllm", "flashinfer_cutlass",
"--moe-runner-backend", "--moe-runner-backend",
cls.backend, cls.backend,
"--tp-size", "--tp-size",