kanshan/TensorRT-LLMs
1
0
Fork 0
mirror of https://github.com/NVIDIA/TensorRT-LLM.git synced 2026-02-18 00:35:04 +08:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity

[None][ci] skip TestGPTOSS (#7333)

Browse Source 
Signed-off-by: junq <22017000+QiJune@users.noreply.github.com>
This commit is contained in:
QI JUN 2025-08-28 17:01:49 +08:00 committed by GitHub
parent 4541655e5f
commit ae89163368
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 4 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
tests/integration/defs/accuracy/test_llm_api_pytorch.py
View File

@ -2719,6 +2719,7 @@ class TestGPTOSS(LlmapiAccuracyTestHarness):
(True, True),
])
def test_w4_1gpu(self, moe_backend, cuda_graph, overlap_scheduler, mocker):
pytest.skip("https://nvbugs/5481087")
if moe_backend == "TRITON" and not IS_TRITON_KERNELS_AVAILABLE:
pytest.skip("Triton kernels are not available")
@ -2736,7 +2737,7 @@ class TestGPTOSS(LlmapiAccuracyTestHarness):
with llm:
model_name = "GPT-OSS/MXFP4"
mocker.patch.object(GSM8K, {"MAX_OUTPUT_LEN": 8192})
mocker.patch.object(GSM8K, "MAX_OUTPUT_LEN", 8192)
task = GSM8K(model_name)
task.evaluate(llm,
extra_evaluator_kwargs=self.extra_evaluator_kwargs)
@ -2756,6 +2757,7 @@ class TestGPTOSS(LlmapiAccuracyTestHarness):
ids=["tp4", "ep4", "dp4"])
def test_w4_4gpus(self, moe_backend, tp_size, pp_size, ep_size,
attention_dp, cuda_graph, overlap_scheduler, mocker):
pytest.skip("https://nvbugs/5481087")
if moe_backend == "TRITON":
if not IS_TRITON_KERNELS_AVAILABLE:
pytest.skip("Triton kernels are not available")
@ -2776,7 +2778,7 @@ class TestGPTOSS(LlmapiAccuracyTestHarness):
with llm:
model_name = "GPT-OSS/MXFP4"
task = GSM8K(model_name)
mocker.patch.object(GSM8K, {"MAX_OUTPUT_LEN": 8192})
mocker.patch.object(GSM8K, "MAX_OUTPUT_LEN", 8192)
task.evaluate(llm,
extra_evaluator_kwargs=self.extra_evaluator_kwargs)