kanshan/TensorRT-LLMs
1
0
Fork 0
mirror of https://github.com/NVIDIA/TensorRT-LLM.git synced 2026-02-18 08:45:05 +08:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity

[None][ci] move some slow test cases of DGX-B200 to post merge (#9467)

Browse Source 
Signed-off-by: junq <22017000+QiJune@users.noreply.github.com>
This commit is contained in:
QI JUN 2025-11-26 10:48:53 +08:00 committed by GitHub
parent 6a64cb4c71
commit 5972119e1c
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 6 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

14
tests/integration/test_lists/test-db/l0_dgx_b200.yml
View File

@ -15,12 +15,10 @@ l0_dgx_b200:
backend: pytorch
orchestrator: mpi
tests:
- unittest/_torch/multi_gpu_modeling -k "deepseek"
- unittest/_torch/modules/test_fused_moe.py::test_fused_moe_alltoall_fp4[DeepEPLowLatency]
- unittest/_torch/modules/test_fused_moe.py::test_fused_moe_alltoall_fp4[MNNVL]
- accuracy/test_llm_api_pytorch.py::TestLlama3_1_8BInstruct::test_bfloat16_4gpus[pp4-attn_backend=TRTLLM-torch_compile=False]
- accuracy/test_llm_api_pytorch.py::TestLlama3_1_8BInstruct::test_fp8_4gpus[tp4-fp8kv=True-attn_backend=TRTLLM-torch_compile=False]
- accuracy/test_llm_api_pytorch.py::TestLlama3_1_8BInstruct::test_fp8_4gpus[pp4-fp8kv=True-attn_backend=TRTLLM-torch_compile=False]
- accuracy/test_llm_api_pytorch.py::TestDeepSeekV3Lite::test_bfloat16_4gpus[tp4-mtp_nextn=2-attention_dp=True-cuda_graph=True-overlap_scheduler=True-torch_compile=True]
- accuracy/test_llm_api_pytorch.py::TestDeepSeekV3Lite::test_bfloat16_4gpus[ep4-mtp_nextn=2-attention_dp=True-cuda_graph=True-overlap_scheduler=True-torch_compile=False]
- accuracy/test_llm_api_pytorch.py::TestDeepSeekV3Lite::test_bfloat16_4gpus[tp2pp2-mtp_nextn=0-attention_dp=False-cuda_graph=False-overlap_scheduler=False-torch_compile=True]
@ -36,7 +34,6 @@ l0_dgx_b200:
- accuracy/test_llm_api_pytorch.py::TestDeepSeekV3Lite::test_nvfp4_4gpus[moe_backend=CUTLASS-mtp_nextn=2-tp2pp2-fp8kv=False-attention_dp=False-cuda_graph=False-overlap_scheduler=False-torch_compile=False]
- accuracy/test_llm_api_pytorch.py::TestDeepSeekV3Lite::test_nvfp4_4gpus[moe_backend=CUTLASS-mtp_nextn=2-pp4-fp8kv=True-attention_dp=True-cuda_graph=True-overlap_scheduler=True-torch_compile=False]
- accuracy/test_llm_api_pytorch.py::TestDeepSeekV3Lite::test_nvfp4_4gpus[moe_backend=TRTLLM-mtp_nextn=2-ep4-fp8kv=True-attention_dp=True-cuda_graph=True-overlap_scheduler=True-torch_compile=False]
- accuracy/test_llm_api_pytorch.py::TestDeepSeekV3Lite::test_nvfp4_4gpus[moe_backend=CUTEDSL-mtp_nextn=2-ep4-fp8kv=True-attention_dp=True-cuda_graph=True-overlap_scheduler=True-torch_compile=False]
- accuracy/test_llm_api_pytorch.py::TestQwen3_30B_A3B::test_nvfp4[tep4_latency_moe_trtllm-torch_compile=True]
- accuracy/test_llm_api_pytorch.py::TestQwen3_30B_A3B::test_nvfp4[dep4_latency_moe_trtllm-torch_compile=False]
- accuracy/test_llm_api_pytorch.py::TestQwen3_30B_A3B::test_nvfp4[dep4_latency_moe_cutlass-torch_compile=False]
@ -52,11 +49,8 @@ l0_dgx_b200:
- accuracy/test_llm_api_pytorch.py::TestGPTOSS::test_w4a16[dp4-fp8]
- accuracy/test_llm_api_pytorch.py::TestGPTOSS::test_eagle3[trtllm-two_model-overlap_scheduler]
- accuracy/test_llm_api_pytorch.py::TestGPTOSS::test_eagle3[cutlass-one_model-overlap_scheduler]
- disaggregated/test_disaggregated.py::test_disaggregated_benchmark_on_diff_backends[DeepSeek-V3-Lite-bf16]
- disaggregated/test_disaggregated.py::test_disaggregated_benchmark_on_diff_backends[llama-3.1-8b-instruct-hf-fp8]
- disaggregated/test_disaggregated.py::test_disaggregated_deepseek_v3_lite_fp8_nixl[DeepSeek-V3-Lite-fp8]
- accuracy/test_llm_api_pytorch.py::TestLlama3_3_70BInstruct::test_fp8_tp4
- accuracy/test_llm_api_pytorch.py::TestLlama3_3_70BInstruct::test_fp4_tp2pp2
- accuracy/test_llm_api_pytorch.py::TestLlama3_3_70BInstruct::test_nvfp4_tp4
- accuracy/test_llm_api_pytorch.py::TestGPTOSS::test_w4_4gpus[dp4-cutlass-auto]
- accuracy/test_llm_api_pytorch.py::TestDeepSeekR1::test_nvfp4_multi_gpus[latency_adp_lmtp_tp4]
@ -198,10 +192,14 @@ l0_dgx_b200:
- accuracy/test_llm_api_pytorch.py::TestGPTOSS::test_w4_4gpus[dp4-triton-auto]
- accuracy/test_llm_api_pytorch.py::TestGPTOSS::test_eagle3[trtllm-one_model-overlap_scheduler]
- accuracy/test_llm_api_pytorch.py::TestGPTOSS::test_eagle3[trtllm-one_model-no_overlap_scheduler]
- accuracy/test_llm_api_pytorch.py::TestGPTOSS::test_eagle3[trtllm-two_model-overlap_scheduler]
- accuracy/test_llm_api_pytorch.py::TestGPTOSS::test_eagle3[trtllm-two_model-no_overlap_scheduler]
- accuracy/test_llm_api_pytorch.py::TestGPTOSS::test_eagle3[cutlass-one_model-overlap_scheduler]
- accuracy/test_llm_api_pytorch.py::TestGPTOSS::test_eagle3[cutlass-one_model-no_overlap_scheduler]
- accuracy/test_llm_api_pytorch.py::TestGPTOSS::test_eagle3[cutlass-two_model-overlap_scheduler]
- accuracy/test_llm_api_pytorch.py::TestGPTOSS::test_eagle3[cutlass-two_model-no_overlap_scheduler]
- disaggregated/test_disaggregated.py::test_disaggregated_benchmark_on_diff_backends[llama-v3-8b-hf]
- disaggregated/test_disaggregated.py::test_disaggregated_benchmark_on_diff_backends[DeepSeek-V3-Lite-bf16]
- disaggregated/test_disaggregated.py::test_disaggregated_benchmark_on_diff_backends[llama-3.1-8b-instruct-hf-fp8]
- unittest/_torch/multi_gpu_modeling -k "deepseek"
- accuracy/test_llm_api_pytorch.py::TestLlama3_1_8BInstruct::test_fp8_4gpus[pp4-fp8kv=True-attn_backend=TRTLLM-torch_compile=False]
- accuracy/test_llm_api_pytorch.py::TestDeepSeekV3Lite::test_nvfp4_4gpus[moe_backend=CUTEDSL-mtp_nextn=2-ep4-fp8kv=True-attention_dp=True-cuda_graph=True-overlap_scheduler=True-torch_compile=False]
- accuracy/test_llm_api_pytorch.py::TestLlama3_3_70BInstruct::test_fp4_tp2pp2