kanshan/TensorRT-LLMs
1
0
Fork 0
mirror of https://github.com/NVIDIA/TensorRT-LLM.git synced 2026-02-16 15:55:08 +08:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity

[https://nvbugs/5804146][fix] Enable responses tests and remove ds to… (#10925)

Browse Source 
Signed-off-by: Junyi Xu <219237550+JunyiXu-nv@users.noreply.github.com>
Signed-off-by: Wangshanshan <30051912+dominicshanshan@users.noreply.github.com>
This commit is contained in:
JunyiXu-nv 2026-01-24 16:04:02 +08:00 committed by Yanchao Lu
parent 196d94a419
commit 90ea6c1e09
2 changed files with 5 additions and 13 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
tests/integration/test_lists/waives.txt
View File

@ -227,7 +227,6 @@ examples/test_ray.py::test_ray_disaggregated_serving[tp2] SKIP (https://nvbugs/5
unittest/executor/test_rpc_proxy.py SKIP (https://nvbugs/5605741)
unittest/executor/test_rpc_worker.py SKIP (https://nvbugs/5605741)
cpp/test_e2e.py::test_model[-redrafter-86] SKIP (https://nvbugs/5761642)
test_e2e.py::test_openai_responses SKIP (https://nvbugs/5804146)
triton_server/test_triton.py::test_gpt_gather_logits[gpt-gather-logits] SKIP (https://nvbugs/5766960)
full:sm89/accuracy/test_disaggregated_serving.py::TestLlama3_1_8BInstruct::test_ctx_pp_gen_tp_asymmetric[MMLU-gen_tp=1-ctx_pp=2] SKIP (https://nvbugs/5596337)
test_e2e.py::test_eagle3_output_consistency_4gpus[Qwen3/saved_models_Qwen3-235B-A22B_fp8_hf-Qwen3/qwen3-235B-eagle3] SKIP (https://nvbugs/5685010)

17
tests/unittest/llmapi/apps/_test_openai_responses.py
View File

@ -13,10 +13,7 @@ pytestmark = pytest.mark.threadleak(enabled=False)
@pytest.fixture(scope="module",
params=[
"gpt_oss/gpt-oss-20b", "DeepSeek-R1-Distill-Qwen-1.5B",
"Qwen3/Qwen3-0.6B"
])
params=["gpt_oss/gpt-oss-20b", "Qwen3/Qwen3-0.6B"])
def model(request):
return request.param
@ -35,10 +32,6 @@ def server(model: str, num_postprocess_workers: int):
args = ["--num_postprocess_workers", f"{num_postprocess_workers}"]
if model.startswith("Qwen3"):
args.extend(["--reasoning_parser", "qwen3"])
elif model.startswith("DeepSeek-R1"):
args.extend(["--reasoning_parser", "deepseek-r1"])
if not model.startswith("gpt_oss"):
args.extend(["--tool_parser", "qwen3"])
with RemoteOpenAIServer(model_path, args) as remote_server:
@ -167,8 +160,8 @@ def get_current_weather(location: str, format: str = "celsius") -> dict:
@pytest.mark.asyncio(loop_scope="module")
async def test_tool_calls(client: openai.AsyncOpenAI, model: str):
if model.startswith("DeepSeek-R1"):
pytest.skip("DeepSeek-R1 does not support tool calls")
if model.startswith("Qwen3"):
pytest.skip("Qwen3 tool call is not stable")
tool_get_current_weather = {
"type": "function",
@ -241,8 +234,8 @@ async def test_streaming(client: openai.AsyncOpenAI, model: str):
@pytest.mark.asyncio(loop_scope="module")
async def test_streaming_tool_call(client: openai.AsyncOpenAI, model: str):
if model.startswith("DeepSeek-R1"):
pytest.skip("DeepSeek-R1 does not support tool calls")
if model.startswith("Qwen3"):
pytest.skip("Qwen3 tool call is not stable")
tool_get_current_weather = {
"type": "function",