test: add test cases for 0.19 release (#3608)

* fix test name

Signed-off-by: Ivy Zhang <[email protected]>

* add quickstart test for nemotron-ultra

Signed-off-by: Ivy Zhang <[email protected]>

* add rcca multi-node test case for deepseek-v3

Signed-off-by: Ivy Zhang <[email protected]>

* add rcca info

Signed-off-by: Ivy Zhang <[email protected]>

---------

Signed-off-by: Ivy Zhang <[email protected]>
Signed-off-by: Ivy Zhang <[email protected]>

Author: crazydemo

tests/integration/defs/test_e2e.py

@@ -1517,6 +1517,31 @@ def test_ptq_quickstart_advanced_mtp(llm_root, llm_venv, model_name,
     ])
 
 
+@pytest.mark.skip_less_device_memory(80000)
+@pytest.mark.skip_less_device(8)
+@pytest.mark.parametrize("model_name,model_path", [
+    pytest.param('DeepSeek-V3', 'DeepSeek-V3', marks=skip_pre_hopper),
+])
+def test_ptp_quickstart_advanced_deepseek_v3_2nodes_8gpus(
+        llm_root, llm_venv, model_name, model_path):
+    # "RCCA https://nvbugs/5163844"
+    print(f"Testing {model_name}.")
+    example_root = Path(os.path.join(llm_root, "examples", "pytorch"))
+    llm_venv.run_cmd([
+        str(example_root / "quickstart_advanced.py"),
+        "--enable_overlap_scheduler",
+        "--model_dir",
+        f"{llm_models_root()}/{model_path}",
+        "--moe_ep_size=8",
+        "--tp_size=16",
+        "--use_cuda_graph",
+        "--kv_cache_fraction=0.5",
+        "--max_batch_size=32",
+        "--max_num_tokens=2048",
+        "--kv_cache_enable_block_reuse",
+    ])
+
+
 @pytest.mark.parametrize("model_name,model_path,eagle_model_path", [
     ("Llama-3.1-8b-Instruct", "llama-3.1-model/Llama-3.1-8B-Instruct",
      "EAGLE3-LLaMA3.1-Instruct-8B"),
@@ -1580,6 +1605,9 @@ def test_ptp_quickstart_advanced_deepseek_r1_8gpus(llm_root, llm_venv,
     pytest.param('Mixtral-8x7B-NVFP4',
                  'nvfp4-quantized/Mixtral-8x7B-Instruct-v0.1',
                  marks=skip_pre_blackwell),
+    pytest.param('Nemotron-Ultra-253B',
+                 'nemotron-nas/Llama-3_1-Nemotron-Ultra-253B-v1',
+                 marks=skip_pre_hopper),
 ])
 def test_ptp_quickstart_advanced_8gpus(llm_root, llm_venv, model_name,
                                        model_path):

tests/integration/test_lists/qa/examples_test_list.txt

@@ -425,14 +425,14 @@ accuracy/test_llm_api.py::TestQwen2_5_7BInstruct::test_fp8
 accuracy/test_llm_api_pytorch.py::TestLlama3_1_8B::test_nvfp4
 accuracy/test_llm_api_pytorch.py::TestLlama3_3_70BInstruct::test_fp8_tp4
 accuracy/test_llm_api_pytorch.py::TestLlama3_3_70BInstruct::test_nvfp4_tp4
-accuracy/test_llm_api_pytorch.py::TestMistral_7B::test_auto_dtype
+accuracy/test_llm_api_pytorch.py::TestMistral7B::test_auto_dtype
 accuracy/test_llm_api_pytorch.py::TestMixtral8x7B::test_fp8_tp2
 accuracy/test_llm_api_pytorch.py::TestMixtral8x7B::test_nvfp4_tp2
 accuracy/test_llm_api_pytorch.py::TestDeepSeekV3Lite::test_bfloat16[]
 accuracy/test_llm_api_pytorch.py::TestDeepSeekV3Lite::test_fp8_block_scales[]
 accuracy/test_llm_api_pytorch.py::TestDeepSeekV3Lite::test_nvfp4[]
 accuracy/test_llm_api_pytorch.py::TestMinitron4BBaseInstruct::test_fp8_prequantized
-accuracy/test_llm_api_pytorch.py::TestNemotronNas::test_auto_dtype
+accuracy/test_llm_api_pytorch.py::TestNemotronNas::test_auto_dtype_tp8
 accuracy/test_llm_api_pytorch.py::TestQwen2_7BInstruct::test_auto_dtype
 
 test_e2e.py::test_benchmark_sanity[bert_base] # 127.18s
@@ -465,6 +465,7 @@ test_e2e.py::test_ptp_quickstart_advanced_8gpus[Llama3.1-70B-FP8-llama-3.1-model
 test_e2e.py::test_ptp_quickstart_advanced_8gpus[Llama3.1-405B-FP8-llama-3.1-model/Llama-3.1-405B-Instruct-FP8]
 test_e2e.py::test_ptp_quickstart_advanced_8gpus[Mixtral-8x7B-BF16-Mixtral-8x7B-v0.1]
 test_e2e.py::test_ptp_quickstart_advanced_8gpus[Mixtral-8x7B-NVFP4-nvfp4-quantized/Mixtral-8x7B-Instruct-v0.1]
+test_e2e.py::test_ptp_quickstart_advanced_8gpus[Nemotron-Ultra-253B-nemotron-nas/Llama-3_1-Nemotron-Ultra-253B-v1]
 test_e2e.py::test_ptp_quickstart_advanced_deepseek_r1_8gpus[DeepSeek-R1-DeepSeek-R1/DeepSeek-R1]
 test_e2e.py::test_ptp_quickstart_multimodal[NVILA-8B-FP16-vila/NVILA-8B-image]
 test_e2e.py::test_ptp_quickstart_multimodal[NVILA-8B-FP16-vila/NVILA-8B-video]

tests/integration/test_lists/qa/llm_multinodes_function_test.txt

@@ -12,5 +12,6 @@ examples/test_llama.py::test_llm_llama_v3_1_2nodes_8gpus[llama-3.1-405b-fp8-disa
 examples/test_llama.py::test_llm_llama_v3_1_2nodes_8gpus[llama-3.1-405b-fp8-disable_fp8-tp8pp2-infer]
 examples/test_mixtral.py::test_llm_mixtral_2nodes_8gpus[Mixtral-8x22B-v0.1-plugin-renormalize-tensor_parallel-build]
 examples/test_mixtral.py::test_llm_mixtral_2nodes_8gpus[Mixtral-8x22B-v0.1-plugin-renormalize-tensor_parallel-infer]
+test_e2e.py::test_ptp_quickstart_advanced_deepseek_v3_2nodes_8gpus[DeepSeek-V3-DeepSeek-V3]
 test_e2e.py::test_openai_multinodes_chat_tp16pp1
 test_e2e.py::test_openai_multinodes_chat_tp8pp2