Skip to content

[https://nvbugs/6273846][test] gate GPT-OSS TRTLLM Gen MoE tests to SM100/SM103 #15128

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
longlee0622 merged 1 commit into from
Jun 22, 2026
+9 −6
Merged

[https://nvbugs/6273846][test] gate GPT-OSS TRTLLM Gen MoE tests to SM100/SM103 #15128

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
15 changes: 9 additions & 6 deletions tests/integration/defs/accuracy/test_llm_api_pytorch.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -5039,13 +5039,16 @@ class TestGPTOSS(LlmapiAccuracyTestHarness):
}

MODEL_PATH = f"{llm_models_root()}/gpt_oss/gpt-oss-120b"
skip_no_trtllm_gen_moe_support = pytest.mark.skipif(
get_sm_version() not in (100, 103),
reason="TRTLLM Gen MoE supports SM100 and SM103 only")

@pytest.mark.parametrize(
"kv_cache_dtype",
["auto", pytest.param("fp8", marks=skip_pre_blackwell)])
@pytest.mark.parametrize("moe_backend", [
"CUTLASS",
pytest.param("TRTLLM", marks=skip_pre_blackwell),
pytest.param("TRTLLM", marks=skip_no_trtllm_gen_moe_support),
pytest.param("TRITON", marks=[skip_no_hopper, skip_no_mxfp4_swizzle])
],
ids=["cutlass", "trtllm", "triton"])
Expand Down Expand Up @@ -5165,7 +5168,7 @@ def test_dflash(self, mocker):
["auto", pytest.param("fp8", marks=skip_pre_blackwell)])
@pytest.mark.parametrize("moe_backend", [
"CUTLASS",
pytest.param("TRTLLM", marks=skip_pre_blackwell),
pytest.param("TRTLLM", marks=skip_no_trtllm_gen_moe_support),
pytest.param("TRITON", marks=[skip_no_hopper, skip_no_mxfp4_swizzle])
],
ids=["cutlass", "trtllm", "triton"])
Expand Down Expand Up @@ -5283,7 +5286,7 @@ def test_w4a16(self, kv_cache_dtype, tp_size, pp_size, ep_size,
["auto", pytest.param("fp8", marks=skip_pre_blackwell)])
@pytest.mark.parametrize("moe_backend", [
"CUTLASS",
pytest.param("TRTLLM", marks=skip_pre_blackwell),
pytest.param("TRTLLM", marks=skip_no_trtllm_gen_moe_support),
pytest.param("TRITON", marks=skip_no_hopper)
],
ids=["cutlass", "trtllm", "triton"])
Expand Down Expand Up @@ -5329,7 +5332,7 @@ def test_w4_2gpus(self, kv_cache_dtype, moe_backend, tp_size, pp_size,
["auto", pytest.param("fp8", marks=skip_pre_blackwell)])
@pytest.mark.parametrize("moe_backend", [
"CUTLASS",
pytest.param("TRTLLM", marks=skip_pre_blackwell),
pytest.param("TRTLLM", marks=skip_no_trtllm_gen_moe_support),
pytest.param("TRITON", marks=skip_no_hopper)
],
ids=["cutlass", "trtllm", "triton"])
Expand Down Expand Up @@ -5395,7 +5398,7 @@ def test_w4_chunked_prefill(self, kv_cache_dtype, moe_backend, mocker):
ids=["one_model", "two_model"])
@pytest.mark.parametrize("moe_backend", [
"CUTLASS",
pytest.param("TRTLLM", marks=skip_pre_blackwell),
pytest.param("TRTLLM", marks=skip_no_trtllm_gen_moe_support),
pytest.param("TRITON", marks=skip_no_hopper)
],
ids=["cutlass", "trtllm", "triton"])
Expand Down Expand Up @@ -5589,7 +5592,7 @@ def test_eagle3_guided_decoding_4gpus(self, one_model, mocker):
ids=["one_model", "two_model"])
@pytest.mark.parametrize("moe_backend", [
"CUTLASS",
pytest.param("TRTLLM", marks=skip_pre_blackwell),
pytest.param("TRTLLM", marks=skip_no_trtllm_gen_moe_support),
pytest.param("TRITON", marks=skip_no_hopper)
],
ids=["cutlass", "trtllm", "triton"])
Expand Down
Loading