diff --git a/AGENTS.md b/AGENTS.md index cf6c717d0928..51853d3a963b 100644 --- a/AGENTS.md +++ b/AGENTS.md @@ -167,7 +167,6 @@ See [CI overview](docs/source/developer-guide/ci-overview.md) for full details. CI is triggered by posting comments on the PR. Basic commands: - `/bot run` — trigger the standard CI pipeline - `/bot run --disable-fail-fast` — run all stages even if earlier ones fail (only add when explicitly needed) -- `/bot run --extra-stage "DGX_B200-4_GPUs-AutoDeploy-1, DGX_H100-4_GPUs-AutoDeploy-1"` — include AutoDeploy CI stages (use for AutoDeploy-related PRs) For a full list of up-to-date bot commands, post `/bot help` as a PR comment and check the bot's reply. diff --git a/jenkins/L0_MergeRequest.groovy b/jenkins/L0_MergeRequest.groovy index aeb28c56a585..6895996cb8b3 100644 --- a/jenkins/L0_MergeRequest.groovy +++ b/jenkins/L0_MergeRequest.groovy @@ -1248,7 +1248,6 @@ def getMultiGpuFileChanged(pipeline, testFilter, globalVars) "tensorrt_llm/_torch/pyexecutor/model_engine.py", "tensorrt_llm/_torch/pyexecutor/py_executor.py", "tensorrt_llm/_torch/weight_sharing/", - "tensorrt_llm/_torch/auto_deploy/transform/library/sharding.py", "tensorrt_llm/_torch/visual_gen/attention_backend/parallel.py", "tensorrt_llm/_torch/visual_gen/modules/vae/", "tensorrt_llm/_torch/visual_gen/modules/attention.py", @@ -1313,7 +1312,6 @@ def getMultiGpuFileChanged(pipeline, testFilter, globalVars) "tests/integration/test_lists/test-db/l0_model_express.yml", "tests/integration/test_lists/test-db/l0_rtx_pro_6000.yml", "tests/integration/test_lists/test-db/l0_verl.yml", - "tests/unittest/auto_deploy/multigpu", "tests/unittest/_torch/multi_gpu/", "tests/unittest/_torch/multi_gpu_modeling/", "tests/unittest/_torch/visual_gen/multi_gpu/", @@ -1323,7 +1321,6 @@ def getMultiGpuFileChanged(pipeline, testFilter, globalVars) "tests/integration/defs/accuracy/test_disaggregated_serving.py", "tests/unittest/_torch/ray_orchestrator/multi_gpu/", "tests/integration/defs/examples/test_ray.py", - "tests/integration/defs/accuracy/test_llm_api_autodeploy.py", "tests/unittest/llmapi/test_async_llm.py", "docker/common/install_ucx.sh", "docker/common/install_nixl.sh", diff --git a/jenkins/L0_Test.groovy b/jenkins/L0_Test.groovy index ac424cf17aee..3eb41f075c15 100644 --- a/jenkins/L0_Test.groovy +++ b/jenkins/L0_Test.groovy @@ -4033,10 +4033,6 @@ def getMakoArgsFromStageName(stageName, parseSysinfo=false) { // If stageName contains "-FMHA-", add "backend=fmha" to makoArgs // At this point, only tests with backend=fmha or unspecified backend will be run makoArgs += ["backend=fmha"] - } else if (stageName.contains("-AutoDeploy-")) { - // If stageName contains "-AutoDeploy-", add "backend=autodeploy" to makoArgs - // At this point, only tests with backend=autodeploy or unspecified backend will be run - makoArgs += ["backend=autodeploy"] } else if (stageName.contains("-Generic-")) { // Generic stages select tests by marker expression rather than backend ownership. makoArgs += ["backend=generic"] @@ -4045,7 +4041,7 @@ def getMakoArgsFromStageName(stageName, parseSysinfo=false) { // At this point, only tests with backend=verl or unspecified backend will be run makoArgs += ["backend=verl"] } else { - // If stageName does not contain "-PyTorch-", "-CPP-", "-Triton-", "-FMHA-", "-AutoDeploy-", or "-Verl-", do not add any backend + // If stageName does not contain "-PyTorch-", "-CPP-", "-Triton-", "-FMHA-", or "-Verl-", do not add any backend // At this point, all tests will be run // For cases where backend is not specified in makoArgs, we will match all types of backends and tests without specified backend } @@ -5863,12 +5859,10 @@ def launchTestJobs(pipeline, testFilter, globalVars) "A30-PyTorch-1": ["a30", "l0_a30", 1, 2], "A30-PyTorch-2": ["a30", "l0_a30", 2, 2], "A30-CPP-1": ["a30", "l0_a30", 1, 1], - "A30-AutoDeploy-1": ["a30", "l0_a30", 1, 1], "A100X-PyTorch-1": ["a100x", "l0_a100", 1, 1], "L40S-PyTorch-1": ["l40s", "l0_l40s", 1, 2], "L40S-PyTorch-2": ["l40s", "l0_l40s", 2, 2], "H100_PCIe-PyTorch-Ray-1": ["h100-cr", "l0_h100", 1, 1], - "H100_PCIe-AutoDeploy-1": ["h100-cr", "l0_h100", 1, 1], "H100_PCIe-CPP-1": ["h100-cr", "l0_h100", 1, 1], // platform, test DB, split, splits, GPU count, ModelExpress sidecars "DGX_H100-2_GPUs-PyTorch-ModelExpress-1": ["dgx-h100-x4", "l0_model_express", 1, 1, 2, true], @@ -5890,7 +5884,6 @@ def launchTestJobs(pipeline, testFilter, globalVars) "A100X-PyTorch-Post-Merge-1": ["a100x", "l0_a100", 1, 1], "L40S-PyTorch-Post-Merge-1": ["l40s", "l0_l40s", 1, 1], "L40S-FMHA-Post-Merge-1": ["l40s", "l0_l40s", 1, 1], - "H100_PCIe-AutoDeploy-Post-Merge-1": ["h100-cr", "l0_h100", 1, 1], "H100_PCIe-FMHA-Post-Merge-1": ["h100-cr", "l0_h100", 1, 1], "H100_PCIe-PyTorch-Perf-1": ["h100-cr", "l0_perf", 1, 1], "DGX_H200-8_GPUs-PyTorch-Post-Merge-1": ["dgx-h200-x8", "l0_dgx_h200", 1, 1, 8], @@ -5940,8 +5933,6 @@ def launchTestJobs(pipeline, testFilter, globalVars) "DGX_H100-4_GPUs-PyTorch-Others-1": ["auto:dgx-h100-x4", "l0_dgx_h100", 1, 2, 4], "DGX_H100-4_GPUs-PyTorch-Others-2": ["auto:dgx-h100-x4", "l0_dgx_h100", 2, 2, 4], "DGX_H100-4_GPUs-PyTorch-Ray-1": ["auto:dgx-h100-x4", "l0_dgx_h100", 1, 1, 4], - "DGX_H100-4_GPUs-AutoDeploy-1": ["auto:dgx-h100-x4", "l0_dgx_h100", 1, 1, 4], - "DGX_H100-4_GPUs-AutoDeploy-Post-Merge-1": ["auto:dgx-h100-x4", "l0_dgx_h100", 1, 1, 4], "DGX_H100-4_GPUs-PyTorch-Post-Merge-1": ["auto:dgx-h100-x4", "l0_dgx_h100", 1, 1, 4], "DGX_B200-PyTorch-1": ["auto:dgx-b200-flex", "l0_b200", 1, 9, 1, 1, true], "DGX_B200-PyTorch-2": ["auto:dgx-b200-flex", "l0_b200", 2, 9, 1, 1, true], @@ -5952,8 +5943,6 @@ def launchTestJobs(pipeline, testFilter, globalVars) "DGX_B200-PyTorch-7": ["auto:dgx-b200-flex", "l0_b200", 7, 9, 1, 1, true], "DGX_B200-PyTorch-8": ["auto:dgx-b200-flex", "l0_b200", 8, 9, 1, 1, true], "DGX_B200-PyTorch-9": ["auto:dgx-b200-flex", "l0_b200", 9, 9, 1, 1, true], - "DGX_B200-AutoDeploy-1": ["auto:dgx-b200-flex", "l0_b200", 1, 1, 1, 1, true], - "DGX_B200-AutoDeploy-Post-Merge-1": ["auto:dgx-b200-flex", "l0_b200", 1, 1, 1, 1, true], "DGX_B200-PyTorch-Post-Merge-1": ["auto:dgx-b200-flex", "l0_b200", 1, 2, 1, 1, true], "DGX_B200-PyTorch-Post-Merge-2": ["auto:dgx-b200-flex", "l0_b200", 2, 2, 1, 1, true], "DGX_B200-2_GPUs-PyTorch-1": ["auto:dgx-b200-flex", "l0_dgx_b200", 1, 1, 2, 1, true], @@ -5961,8 +5950,6 @@ def launchTestJobs(pipeline, testFilter, globalVars) "DGX_B200-4_GPUs-PyTorch-2": ["auto:dgx-b200-flex", "l0_dgx_b200", 2, 3, 4, 1, true], "DGX_B200-4_GPUs-PyTorch-3": ["auto:dgx-b200-flex", "l0_dgx_b200", 3, 3, 4, 1, true], "DGX_B200-4_GPUs-PyTorch-Ray-1": ["auto:dgx-b200-flex", "l0_dgx_b200", 1, 1, 4, 1, true], - "DGX_B200-4_GPUs-AutoDeploy-1": ["auto:dgx-b200-flex", "l0_dgx_b200", 1, 1, 4, 1, true], - "DGX_B200-4_GPUs-AutoDeploy-Post-Merge-1": ["auto:dgx-b200-flex", "l0_dgx_b200", 1, 1, 4, 1, true], "DGX_B200-4_GPUs-PyTorch-Post-Merge-1": ["auto:dgx-b200-flex", "l0_dgx_b200", 1, 4, 4, 1, true], "DGX_B200-4_GPUs-PyTorch-Post-Merge-2": ["auto:dgx-b200-flex", "l0_dgx_b200", 2, 4, 4, 1, true], "DGX_B200-4_GPUs-PyTorch-Post-Merge-3": ["auto:dgx-b200-flex", "l0_dgx_b200", 3, 4, 4, 1, true], @@ -5972,7 +5959,6 @@ def launchTestJobs(pipeline, testFilter, globalVars) "DGX_B200-8_GPUs-PyTorch-3": ["auto:dgx-b200-flex", "l0_dgx_b200", 3, 4, 8, 1, true], "DGX_B200-8_GPUs-PyTorch-4": ["auto:dgx-b200-flex", "l0_dgx_b200", 4, 4, 8, 1, true], "DGX_B200-8_GPUs-PyTorch-Ray-1": ["auto:dgx-b200-flex", "l0_dgx_b200", 1, 1, 8, 1, true], - "DGX_B200-8_GPUs-AutoDeploy-Post-Merge-1": ["auto:dgx-b200-flex", "l0_dgx_b200", 1, 1, 8, 1, true], "DGX_B200-4_GPUs-Verl-Post-Merge-1": ["auto:dgx-b200-flex", "l0_verl", 1, 1, 4, 1, true], "B300-PyTorch-1": ["auto:dgx-b300-flex", "l0_b300", 1, 2, 1, 1, true], "B300-PyTorch-2": ["auto:dgx-b300-flex", "l0_b300", 2, 2, 1, 1, true], diff --git a/tests/integration/test_lists/qa/llm_function_core.txt b/tests/integration/test_lists/qa/llm_function_core.txt index d0e17e71119d..13e465b328d0 100644 --- a/tests/integration/test_lists/qa/llm_function_core.txt +++ b/tests/integration/test_lists/qa/llm_function_core.txt @@ -76,46 +76,6 @@ accuracy/test_dwdp_disaggregated_serving.py::TestDwdpDeepSeekV3Lite::test_dwdp_a accuracy/test_epd_disagg_multimodal.py::TestVideoMMEEPD::test_disaggregated_videomme[nemotron_nano_v3_omni_fp8] accuracy/test_epd_disagg_multimodal.py::TestVideoMMEEPD::test_disaggregated_videomme[nemotron_nano_v3_omni_nvfp4] accuracy/test_epd_disagg_multimodal.py::TestVideoMMEEPD::test_disaggregated_videomme[qwen3vl_2b_instruct] -accuracy/test_llm_api_autodeploy.py::TestLlama3_1_8B::test_attention_dp[4] -accuracy/test_llm_api_autodeploy.py::TestLlama3_1_8B::test_auto_dtype[torch-True-1] -accuracy/test_llm_api_autodeploy.py::TestLlama3_1_8B::test_auto_dtype[triton-False-1] -accuracy/test_llm_api_autodeploy.py::TestLlama3_1_8B::test_auto_dtype[trtllm-False-1] -accuracy/test_llm_api_autodeploy.py::TestLlama3_1_8B::test_auto_dtype[trtllm-False-4] -accuracy/test_llm_api_autodeploy.py::TestLlama3_1_8B::test_auto_dtype[trtllm-True-1] -accuracy/test_llm_api_autodeploy.py::TestLlama3_1_8B_Instruct_Eagle3::test_eagle3_one_model[flashinfer] -accuracy/test_llm_api_autodeploy.py::TestLlama3_1_8B_Instruct_Eagle3::test_eagle3_one_model[trtllm] -accuracy/test_llm_api_autodeploy.py::TestModelRegistryAccuracy::test_autodeploy_from_registry[Qwen_QwQ-32B-False] -accuracy/test_llm_api_autodeploy.py::TestModelRegistryAccuracy::test_autodeploy_from_registry[deepseek-ai_DeepSeek-R1-0528-True] -accuracy/test_llm_api_autodeploy.py::TestModelRegistryAccuracy::test_autodeploy_from_registry[google_gemma-3-1b-it-False] -accuracy/test_llm_api_autodeploy.py::TestModelRegistryAccuracy::test_autodeploy_from_registry[meta-llama_Llama-3.1-8B-Instruct-False] -accuracy/test_llm_api_autodeploy.py::TestModelRegistryAccuracy::test_autodeploy_from_registry[meta-llama_Llama-3.3-70B-Instruct-False] -accuracy/test_llm_api_autodeploy.py::TestModelRegistryAccuracy::test_autodeploy_from_registry[mistralai_Ministral-8B-Instruct-2410-False] -accuracy/test_llm_api_autodeploy.py::TestModelRegistryAccuracy::test_autodeploy_from_registry[nvidia_DeepSeek-R1-0528-NVFP4-v2-True] -accuracy/test_llm_api_autodeploy.py::TestModelRegistryAccuracy::test_autodeploy_from_registry[nvidia_Llama-3.1-8B-Instruct-FP8-True] -accuracy/test_llm_api_autodeploy.py::TestModelRegistryAccuracy::test_autodeploy_from_registry[nvidia_Llama-3.1-8B-Instruct-NVFP4-True] -accuracy/test_llm_api_autodeploy.py::TestNemotronNanoV3::test_accuracy[bf16-1-attn_dp_off-trtllm] -accuracy/test_llm_api_autodeploy.py::TestNemotronNanoV3::test_accuracy[bf16-4-attn_dp_off-trtllm] -accuracy/test_llm_api_autodeploy.py::TestNemotronNanoV3::test_accuracy[fp8-1-attn_dp_off-trtllm] -accuracy/test_llm_api_autodeploy.py::TestNemotronNanoV3::test_accuracy[fp8-4-attn_dp_off-trtllm] -accuracy/test_llm_api_autodeploy.py::TestNemotronNanoV3::test_accuracy[nvfp4-1-attn_dp_off-trtllm] -accuracy/test_llm_api_autodeploy.py::TestNemotronNanoV3::test_accuracy[nvfp4-4-attn_dp_off-trtllm] -accuracy/test_llm_api_autodeploy.py::TestNemotronSuperV3::test_accuracy[bf16-4-attn_dp_off-trtllm] -accuracy/test_llm_api_autodeploy.py::TestNemotronSuperV3::test_accuracy[bf16-4-attn_dp_on-trtllm] -accuracy/test_llm_api_autodeploy.py::TestNemotronSuperV3::test_accuracy[fp8-4-attn_dp_off-trtllm] -accuracy/test_llm_api_autodeploy.py::TestNemotronSuperV3::test_accuracy[fp8-4-attn_dp_on-trtllm] -accuracy/test_llm_api_autodeploy.py::TestNemotronSuperV3::test_accuracy[nvfp4-1-attn_dp_off-trtllm] -accuracy/test_llm_api_autodeploy.py::TestNemotronSuperV3::test_accuracy[nvfp4-4-attn_dp_on-trtllm] -accuracy/test_llm_api_autodeploy.py::TestNemotronSuperV3::test_functional_small[bf16] -accuracy/test_llm_api_autodeploy.py::TestNemotronSuperV3::test_functional_small[fp8] -accuracy/test_llm_api_autodeploy.py::TestNemotronSuperV3::test_mtp[bf16_ws4_180gb-flashinfer] -accuracy/test_llm_api_autodeploy.py::TestNemotronSuperV3::test_mtp[bf16_ws4_180gb-trtllm] -accuracy/test_llm_api_autodeploy.py::TestNemotronSuperV3::test_mtp[fp8_ws4_80gb-trtllm] -accuracy/test_llm_api_autodeploy.py::TestNemotronSuperV3::test_mtp[nvfp4_ws4_80gb-trtllm] -accuracy/test_llm_api_autodeploy.py::TestNemotronV2::test_auto_dtype[False] -accuracy/test_llm_api_autodeploy.py::TestNemotronV2::test_auto_dtype[True] -accuracy/test_llm_api_autodeploy.py::TestNemotronV2::test_fp8[True] -accuracy/test_llm_api_autodeploy.py::TestNemotronV2::test_nvfp4[True] -accuracy/test_llm_api_autodeploy.py::TestQwen3_5_397B_MoE::test_nvfp4[8] accuracy/test_llm_api_pytorch.py::TestDeepSeekR1::test_fp8_blockscale[latency] accuracy/test_llm_api_pytorch.py::TestDeepSeekR1::test_fp8_blockscale[throughput] accuracy/test_llm_api_pytorch.py::TestDeepSeekR1::test_fp8_blockscale[throughput_mtp] @@ -804,11 +764,6 @@ accuracy/test_llm_api_pytorch_multimodal.py::TestQwen3_5_35B_A3B_VL::test_auto_d accuracy/test_llm_api_pytorch_multimodal.py::TestQwen3_5_35B_A3B_VL::test_fp8_prequantized accuracy/test_llm_api_pytorch_ray.py::TestLlama3_1_8BInstruct::test_pp2_ray unittest/disaggregated/test_cache_transceiver_single_process.py::test_cache_transceiver_v1_masked_dsa_indexer_across_asymmetric_pp -disaggregated/test_ad_disagg.py::test_async_eagle3_full_model_handoff -disaggregated/test_ad_disagg.py::test_async_generation_matches_aggregate -disaggregated/test_ad_disagg.py::test_async_generation_no_overlap_matches_aggregate -disaggregated/test_ad_disagg.py::test_async_sharded_generation_handoff -disaggregated/test_ad_disagg_trtllm_serve.py::test_openai_completion disaggregated/test_aiperf_gate.py::test_all_cancelled_fails disaggregated/test_aiperf_gate.py::test_corrupt_export_fails disaggregated/test_aiperf_gate.py::test_empty_export_fails diff --git a/tests/integration/test_lists/qa/llm_function_rtx6k.txt b/tests/integration/test_lists/qa/llm_function_rtx6k.txt index 4208bda5c18b..6e01ea8910e2 100644 --- a/tests/integration/test_lists/qa/llm_function_rtx6k.txt +++ b/tests/integration/test_lists/qa/llm_function_rtx6k.txt @@ -8,7 +8,6 @@ accuracy/test_disaggregated_serving.py::TestGPTOSS::test_auto_dtype[False] accuracy/test_disaggregated_serving.py::TestGPTOSS::test_auto_dtype[True] accuracy/test_disaggregated_serving.py::TestQwen3_8B::test_auto_dtype[False-False] accuracy/test_disaggregated_serving.py::TestQwen3_8B::test_auto_dtype[False-True] -accuracy/test_llm_api_autodeploy.py::TestQwen3_5_397B_MoE::test_nvfp4[8] accuracy/test_llm_api_pytorch.py::TestDeepSeekR1::test_nvfp4_multi_gpus_sm120[throughput_tp8] accuracy/test_llm_api_pytorch.py::TestDeepSeekV3Lite::test_bfloat16[mtp_nextn=0-attention_dp=False-cuda_graph=False-overlap_scheduler=False-torch_compile=False-enable_chunked_prefill=False-v2_kv_cache=False] accuracy/test_llm_api_pytorch.py::TestDeepSeekV3Lite::test_bfloat16[mtp_nextn=0-attention_dp=False-cuda_graph=False-overlap_scheduler=False-torch_compile=False-enable_chunked_prefill=False-v2_kv_cache=True] diff --git a/tests/integration/test_lists/test-db/l0_a30.yml b/tests/integration/test_lists/test-db/l0_a30.yml index 3a5374fa3592..b165611714e6 100644 --- a/tests/integration/test_lists/test-db/l0_a30.yml +++ b/tests/integration/test_lists/test-db/l0_a30.yml @@ -63,25 +63,3 @@ l0_a30: - cpp/test_unit_tests.py::test_unit_tests[layers-80] - cpp/test_unit_tests.py::test_unit_tests[runtime-80] - cpp/test_unit_tests.py::test_unit_tests[thop-80] -# ------------- AutoDeploy Backend Stages --------------- -- condition: - ranges: - system_gpu_count: - gte: 1 - lte: 1 - wildcards: - gpu: - - '*a30*' - linux_distribution_name: ubuntu* - terms: - stage: pre_merge - backend: autodeploy - tests: - - unittest/auto_deploy/singlegpu/compile - - unittest/auto_deploy/singlegpu/custom_ops - - unittest/auto_deploy/singlegpu/mlir - - unittest/auto_deploy/singlegpu/models - - unittest/auto_deploy/singlegpu/shim - - unittest/auto_deploy/singlegpu/smoke - - unittest/auto_deploy/singlegpu/transformations - - unittest/auto_deploy/singlegpu/utils diff --git a/tests/integration/test_lists/test-db/l0_b200.yml b/tests/integration/test_lists/test-db/l0_b200.yml index 47c31ce67519..eb98bc2a7b27 100644 --- a/tests/integration/test_lists/test-db/l0_b200.yml +++ b/tests/integration/test_lists/test-db/l0_b200.yml @@ -439,66 +439,3 @@ l0_b200: - unittest/_torch/modules/moe/test_moe_backend.py::test_moe_backend -k "MEGAMOE_DEEPGEMM" # ---- non-quantized (quant=None) moved to post-merge ---- - unittest/_torch/modules/moe/test_moe_module.py::test_configurable_moe_single_gpu -k "CUTLASS and None" -# ------------- AutoDeploy Backend Stages --------------- -- condition: - ranges: - system_gpu_count: - gte: 1 - lte: 1 - wildcards: - gpu: - - '*b100*' - - '*b200*' - linux_distribution_name: ubuntu* - terms: - stage: pre_merge - backend: autodeploy - tests: - - accuracy/test_llm_api_autodeploy.py::TestLlama3_1_8B::test_auto_dtype[trtllm-False-1] - - accuracy/test_llm_api_autodeploy.py::TestNemotronV2::test_fp8[True] - - accuracy/test_llm_api_autodeploy.py::TestNemotronNanoV3::test_accuracy[fp8-1-attn_dp_off-trtllm] - - accuracy/test_llm_api_autodeploy.py::TestNemotronSuperV3::test_accuracy[nvfp4-1-attn_dp_off-trtllm] - - accuracy/test_llm_api_autodeploy.py::TestNemotronSuperV3::test_functional_small[bf16] - # ------------- AutoDeploy unit tests --------------- - # B-tier keeps only Blackwell-meaningful tests (NVFP4 / FP4 / TRTLLM-Gen / - # SM100+-gated). HW-agnostic tests (compile, models, shim, utils, plus most - # custom_ops, smoke, transformations files) and pure-FP8 tests are covered - # on Hopper (l0_h100.yml) and not duplicated here. - - unittest/auto_deploy/singlegpu/custom_ops/attention/test_triton_attention.py::TestSDPADispatch - - unittest/auto_deploy/singlegpu/custom_ops/mamba/test_flashinfer_mamba_cached_op.py - - unittest/auto_deploy/singlegpu/custom_ops/moe/test_ad_moe_op.py - - unittest/auto_deploy/singlegpu/custom_ops/moe/test_trtllm_moe.py - - unittest/auto_deploy/singlegpu/custom_ops/quantization/test_quant.py - - unittest/auto_deploy/singlegpu/smoke/test_ad_build_small_single.py -k "Nemotron-3-Nano-30B-A3B-FP8 or Nemotron-Nano-3-30B-A3.5B-dev or Llama-4-Scout" - - unittest/auto_deploy/singlegpu/smoke/test_ad_speculative_decoding.py - - unittest/auto_deploy/singlegpu/transformations/library/test_fuse_relu2_quant_nvfp4.py - - unittest/auto_deploy/singlegpu/transformations/library/test_moe_fusion.py - - unittest/auto_deploy/singlegpu/transformations/library/test_nvfp4_swiglu.py - - unittest/auto_deploy/singlegpu/transformations/library/test_quant_fusion.py - - unittest/auto_deploy/singlegpu/transformations/library/test_quant_moe.py - - unittest/auto_deploy/singlegpu/transformations/library/test_quantization.py - - unittest/auto_deploy/singlegpu/transformations/library/test_trtllm_gen_diag.py - -- condition: - ranges: - system_gpu_count: - gte: 1 - lte: 1 - wildcards: - gpu: - - '*b100*' - - '*b200*' - linux_distribution_name: ubuntu* - terms: - stage: post_merge - backend: autodeploy - tests: - - accuracy/test_llm_api_autodeploy.py::TestLlama3_1_8B::test_auto_dtype[triton-False-1] - - accuracy/test_llm_api_autodeploy.py::TestLlama3_1_8B::test_auto_dtype[torch-True-1] - - accuracy/test_llm_api_autodeploy.py::TestNemotronV2::test_auto_dtype[False] - - accuracy/test_llm_api_autodeploy.py::TestNemotronV2::test_auto_dtype[True] - - accuracy/test_llm_api_autodeploy.py::TestNemotronV2::test_nvfp4[True] - - accuracy/test_llm_api_autodeploy.py::TestNemotronNanoV3::test_accuracy[nvfp4-1-attn_dp_off-trtllm] - - accuracy/test_llm_api_autodeploy.py::TestNemotronSuperV3::test_functional_small[fp8] - # ------------- AutoDeploy Perf Sanity --------------- - - perf/test_perf_sanity.py::test_e2e[aggr_upload-super_ad_blackwell-super_ad_ws1_1k1k] TIMEOUT (120) diff --git a/tests/integration/test_lists/test-db/l0_cpu.yml b/tests/integration/test_lists/test-db/l0_cpu.yml index 7ceb1c0b8d37..2b46f28c6873 100644 --- a/tests/integration/test_lists/test-db/l0_cpu.yml +++ b/tests/integration/test_lists/test-db/l0_cpu.yml @@ -24,7 +24,6 @@ l0_cpu: - disaggregated/test_aiperf_gate.py::test_was_cancelled_metadata_fallback - disaggregated/test_aiperf_gate.py::test_non_request_records_excluded_from_denominator - unittest/_torch/attention - - unittest/_torch/auto_deploy - unittest/_torch/cute_dsl/test_kimi_k3_kda_ptx_patch.py - unittest/_torch/distributed - unittest/_torch/executor @@ -58,9 +57,6 @@ l0_cpu: - unittest/_torch/visual_gen/test_visual_gen_utils.py - unittest/_torch/weight_sharing - unittest/api_stability - - unittest/auto_deploy/multigpu - - unittest/auto_deploy/singlegpu - - unittest/auto_deploy/standalone - unittest/bindings - unittest/disaggregated - unittest/dynamo diff --git a/tests/integration/test_lists/test-db/l0_dgx_b200.yml b/tests/integration/test_lists/test-db/l0_dgx_b200.yml index 884eb3e26bbf..8291d3ec0102 100644 --- a/tests/integration/test_lists/test-db/l0_dgx_b200.yml +++ b/tests/integration/test_lists/test-db/l0_dgx_b200.yml @@ -364,88 +364,3 @@ l0_dgx_b200: - accuracy/test_llm_api_pytorch.py::TestDeepSeekV32::test_nvfp4_attn_multi_gpus TIMEOUT (60) # ---- FP8 per-tensor (QDQ) moved to post-merge; block-scale/W4A8 stay in pre ---- - unittest/_torch/modules/moe/test_moe_module.py::test_configurable_moe_multi_gpu -k "CUTLASS and FP8 and not FP8_BLOCK_SCALES and not W4A8 and not MXFP8" -# ------------- AutoDeploy Backend Stages --------------- -- condition: - ranges: - system_gpu_count: - gte: 4 - lte: 4 - wildcards: - gpu: - - '*b200*' - linux_distribution_name: ubuntu* - cpu: x86_64 - terms: - stage: pre_merge - backend: autodeploy - orchestrator: mpi - tests: - - unittest/auto_deploy/multigpu/custom_ops - - unittest/auto_deploy/multigpu/smoke - # B-tier keeps the NVFP4 MoE TP sharding test (SM100+ gated) and the - # strategy_auto parametrization of the allreduce-residual-rmsnorm fusion test - # (covers AUTO -> NVLS dispatch on B-tier, structurally different from H100). - # The other multigpu/transformations tests are HW-agnostic and covered on - # Hopper (l0_dgx_h100.yml). - - unittest/auto_deploy/multigpu/transformations/library/test_tp_sharding.py::test_moe_tp_shard_nvfp4 - - unittest/auto_deploy/multigpu/transformations/library/test_allreduce_residual_rmsnorm_fusion.py -k "strategy_auto" - - accuracy/test_llm_api_autodeploy.py::TestLlama3_1_8B::test_auto_dtype[trtllm-False-4] - - accuracy/test_llm_api_autodeploy.py::TestNemotronNanoV3::test_accuracy[fp8-4-attn_dp_off-trtllm] - - accuracy/test_llm_api_autodeploy.py::TestNemotronSuperV3::test_accuracy[fp8-4-attn_dp_off-trtllm] - - accuracy/test_llm_api_autodeploy.py::TestNemotronUltraV3::test_accuracy[nvfp4-4] - - accuracy/test_llm_api_autodeploy.py::TestModelRegistryAccuracy::test_autodeploy_from_registry[nvidia_Llama-3.1-8B-Instruct-NVFP4-True] - - accuracy/test_llm_api_autodeploy.py::TestGPTOSS::test_mxfp4_gsm8k[120b-tp2] - # ------------- AutoDeploy Perf Sanity --------------- - - perf/test_perf_sanity.py::test_e2e[aggr_upload-super_ad_blackwell-super_ad_ws4_1k1k] TIMEOUT (120) -- condition: - ranges: - system_gpu_count: - gte: 4 - lte: 4 - wildcards: - gpu: - - '*b200*' - linux_distribution_name: ubuntu* - cpu: x86_64 - terms: - stage: post_merge - backend: autodeploy - orchestrator: mpi - tests: - # Move to post-merge due to https://nvbugspro.nvidia.com/bug/6221483 - - accuracy/test_llm_api_autodeploy.py::TestNemotronSuperV3::test_mtp[nvfp4_ws4_80gb-trtllm] - - accuracy/test_llm_api_autodeploy.py::TestNemotronNanoV3::test_accuracy[nvfp4-4-attn_dp_off-trtllm] - - accuracy/test_llm_api_autodeploy.py::TestNemotronSuperV3::test_accuracy[bf16-4-attn_dp_off-trtllm] - - accuracy/test_llm_api_autodeploy.py::TestNemotronSuperV3::test_accuracy[fp8-4-attn_dp_on-trtllm] - - accuracy/test_llm_api_autodeploy.py::TestNemotronSuperV3::test_accuracy[nvfp4-4-attn_dp_on-trtllm] - - accuracy/test_llm_api_autodeploy.py::TestNemotronSuperV3::test_mtp[bf16_ws4_180gb-flashinfer] - - accuracy/test_llm_api_autodeploy.py::TestNemotronSuperV3::test_mtp[bf16_ws4_180gb-trtllm] - - accuracy/test_llm_api_autodeploy.py::TestGPTOSS::test_mxfp4_gsm8k[20b] - - accuracy/test_llm_api_autodeploy.py::TestGPTOSS::test_mxfp4_gsm8k[120b] - - accuracy/test_llm_api_autodeploy.py::TestGPTOSS::test_mxfp4_gsm8k[120b-ep2] - # ------------- AutoDeploy Perf Sanity --------------- - - perf/test_perf_sanity.py::test_e2e[aggr_upload-super_mtp_ad_blackwell-super_mtp_ad_ws4_1k1k] TIMEOUT (120) -# ------------- AutoDeploy Post Merge 8 GPU tests --------------- -- condition: - ranges: - system_gpu_count: - gte: 8 - lte: 8 - wildcards: - gpu: - - '*b200*' - linux_distribution_name: ubuntu* - cpu: x86_64 - terms: - stage: post_merge - backend: autodeploy - auto_trigger: others - orchestrator: mpi - tests: - - accuracy/test_llm_api_autodeploy.py::TestModelRegistryAccuracy::test_autodeploy_from_registry[deepseek-ai_DeepSeek-R1-0528-True] - - accuracy/test_llm_api_autodeploy.py::TestModelRegistryAccuracy::test_autodeploy_from_registry[nvidia_DeepSeek-R1-0528-NVFP4-v2-True] - - accuracy/test_llm_api_autodeploy.py::TestQwen3_5_397B_MoE::test_nvfp4[8] - - accuracy/test_llm_api_autodeploy.py::TestNemotronUltraV3::test_accuracy[nvfp4-8] - - accuracy/test_llm_api_autodeploy.py::TestNemotronSuperV3::test_mtp[nvfp4_ws8_80gb-trtllm] - # ------------- AutoDeploy Perf Sanity --------------- - - perf/test_perf_sanity.py::test_e2e[aggr_upload-deepseek_r1_fp8_ad_blackwell-r1_fp8_ad_ws8_1k1k] TIMEOUT (120) diff --git a/tests/integration/test_lists/test-db/l0_dgx_h100.yml b/tests/integration/test_lists/test-db/l0_dgx_h100.yml index 5077cec41a2c..799fbc321cbd 100644 --- a/tests/integration/test_lists/test-db/l0_dgx_h100.yml +++ b/tests/integration/test_lists/test-db/l0_dgx_h100.yml @@ -321,67 +321,6 @@ l0_dgx_h100: - unittest/llmapi/test_async_llm.py -m "gpu4" - ray_orchestrator/RL/test_rl_perf_reproduce.py::test_rl_perf_reproduce[tp2_2instances] - ray_orchestrator/RL/test_rl_perf_reproduce.py::test_rl_perf_reproduce[tp1_4instances] -# ------------- AutoDeploy Backend Stages --------------- -- condition: - ranges: - system_gpu_count: - gte: 4 - lte: 4 - wildcards: - gpu: - - '*h100*' - linux_distribution_name: ubuntu* - terms: - stage: pre_merge - backend: autodeploy - auto_trigger: others - orchestrator: mpi - tests: - - unittest/auto_deploy/multigpu/compile - - unittest/auto_deploy/multigpu/custom_ops - - unittest/auto_deploy/multigpu/smoke - - unittest/auto_deploy/multigpu/transformations - - accuracy/test_llm_api_autodeploy.py::TestLlama3_1_8B::test_auto_dtype[trtllm-False-4] - - accuracy/test_llm_api_autodeploy.py::TestNemotronNanoV3::test_accuracy[fp8-4-attn_dp_off-trtllm] - - accuracy/test_llm_api_autodeploy.py::TestNemotronNanoV3::test_accuracy[fp8-4-attn_dp_on-trtllm] - - accuracy/test_llm_api_autodeploy.py::TestNemotronSuperV3::test_accuracy[fp8-4-attn_dp_off-trtllm] - - accuracy/test_llm_api_autodeploy.py::TestNemotronSuperV3::test_mtp[fp8_ws4_80gb-trtllm] - - accuracy/test_llm_api_autodeploy.py::TestLlama3_1_8B::test_attention_dp[4] - - accuracy/test_llm_api_autodeploy.py::TestQwen3_5_397B_MoE::test_bf16_small[4] - - accuracy/test_llm_api_autodeploy.py::TestGemma4MoE::test_bf16 - - accuracy/test_llm_api_autodeploy.py::TestMiniMaxM2::test_finegrained_fp8 - - disaggregated/test_ad_disagg.py::test_async_generation_matches_aggregate - - disaggregated/test_ad_disagg.py::test_async_generation_no_overlap_matches_aggregate - - disaggregated/test_ad_disagg.py::test_async_sharded_generation_handoff - - disaggregated/test_ad_disagg.py::test_async_eagle3_full_model_handoff -# ------------- AutoDeploy Backend Stages L1 / Nightly only --------------- -- condition: - ranges: - system_gpu_count: - gte: 4 - lte: 4 - wildcards: - gpu: - - '*h100*' - linux_distribution_name: ubuntu* - terms: - stage: post_merge - backend: autodeploy - auto_trigger: others - orchestrator: mpi - tests: - - disaggregated/test_ad_disagg_trtllm_serve.py::test_openai_completion - - accuracy/test_llm_api_autodeploy.py::TestModelRegistryAccuracy::test_autodeploy_from_registry[google_gemma-3-1b-it-False] - - accuracy/test_llm_api_autodeploy.py::TestModelRegistryAccuracy::test_autodeploy_from_registry[meta-llama_Llama-3.1-8B-Instruct-False] - - accuracy/test_llm_api_autodeploy.py::TestModelRegistryAccuracy::test_autodeploy_from_registry[mistralai_Ministral-8B-Instruct-2410-False] - - accuracy/test_llm_api_autodeploy.py::TestModelRegistryAccuracy::test_autodeploy_from_registry[Qwen_QwQ-32B-False] - - accuracy/test_llm_api_autodeploy.py::TestModelRegistryAccuracy::test_autodeploy_from_registry[meta-llama_Llama-3.3-70B-Instruct-False] - - accuracy/test_llm_api_autodeploy.py::TestModelRegistryAccuracy::test_autodeploy_from_registry[nvidia_Llama-3.1-8B-Instruct-FP8-True] - - accuracy/test_llm_api_autodeploy.py::TestLlama3_1_8B_Instruct_Eagle3::test_eagle3_one_model[flashinfer] - - accuracy/test_llm_api_autodeploy.py::TestNemotronNanoV3::test_accuracy[bf16-4-attn_dp_off-trtllm] - - accuracy/test_llm_api_autodeploy.py::TestNemotronSuperV3::test_accuracy[bf16-4-attn_dp_off-trtllm] - - accuracy/test_llm_api_autodeploy.py::TestNemotronSuperV3::test_accuracy[bf16-4-attn_dp_on-trtllm] - - accuracy/test_llm_api_autodeploy.py::TestNemotronSuperV3::test_accuracy[fp8-4-attn_dp_on-trtllm] - condition: ranges: system_gpu_count: diff --git a/tests/integration/test_lists/test-db/l0_h100.yml b/tests/integration/test_lists/test-db/l0_h100.yml index 1dc9615e0090..5739eaf4116e 100644 --- a/tests/integration/test_lists/test-db/l0_h100.yml +++ b/tests/integration/test_lists/test-db/l0_h100.yml @@ -16,7 +16,6 @@ l0_h100: tests: # ------------- PyTorch tests --------------- - unittest/_torch/attention - - unittest/_torch/auto_deploy - unittest/_torch/compilation - unittest/_torch/debugger - unittest/_torch/executor @@ -469,67 +468,3 @@ l0_h100: backend: fmha tests: - test_fmha.py::test_fmha TIMEOUT (90) -# ------------- AutoDeploy Backend Stages --------------- -- condition: - ranges: - system_gpu_count: - gte: 1 - lte: 1 - wildcards: - gpu: - - '*h100*' - linux_distribution_name: ubuntu* - terms: - stage: pre_merge - backend: autodeploy - orchestrator: mpi - tests: - - unittest/auto_deploy/singlegpu/compile - - unittest/auto_deploy/singlegpu/custom_ops - - unittest/auto_deploy/singlegpu/mlir - - unittest/auto_deploy/singlegpu/models - - unittest/auto_deploy/singlegpu/shim - - unittest/auto_deploy/singlegpu/smoke - - unittest/auto_deploy/singlegpu/transformations - - unittest/auto_deploy/singlegpu/utils - - unittest/auto_deploy/standalone - - accuracy/test_llm_api_autodeploy.py::TestLlama3_1_8B::test_auto_dtype[trtllm-False-1] - - accuracy/test_llm_api_autodeploy.py::TestLlama3_1_8B_Instruct_Eagle3::test_eagle3_one_model[trtllm] - - accuracy/test_llm_api_autodeploy.py::TestGemmaE2B::test_gemma3n_e2b_it - - accuracy/test_llm_api_autodeploy.py::TestGemmaE2B::test_gemma4_e2b_it - - accuracy/test_llm_api_autodeploy.py::TestNemotronV2::test_fp8[True] - - accuracy/test_llm_api_autodeploy.py::TestNemotronNanoV3::test_accuracy[fp8-1-attn_dp_off-trtllm] - - accuracy/test_llm_api_autodeploy.py::TestLlama3_1_8B::test_auto_dtype[triton-False-1] - - examples/test_ad_speculative_decoding.py::test_autodeploy_eagle3_one_model_acceptance_rate[trtllm-torch-cudagraph] - - examples/test_ad_speculative_decoding.py::test_autodeploy_eagle3_one_model_acceptance_rate[flashinfer-torch-simple] - - examples/test_ad_speculative_decoding.py::test_eagle_model_with_weights - - examples/test_ad_speculative_decoding.py::test_eagle_wrapper_forward[1] - - examples/test_ad_speculative_decoding.py::test_eagle_wrapper_forward[2] - - examples/test_ad_speculative_decoding.py::test_nemotron_mtp_model_with_weights - - examples/test_ad_guided_decoding.py::test_autodeploy_guided_decoding_main_json - - disaggregated/test_ad_disagg.py::test_disaggregated_logits[tinyllama] - - disaggregated/test_ad_disagg.py::test_disaggregated_logits[deepseek_v3_mla] - - disaggregated/test_ad_disagg.py::test_reduced_layer_handoff_matches_aggregate[tinyllama] - - disaggregated/test_ad_disagg.py::test_reduced_layer_handoff_matches_aggregate[deepseek_v3_mla] - - disaggregated/test_ad_disagg.py::test_tinyllama_batch_handoff_semantic_slots - - disaggregated/test_ad_disagg.py::test_chunked_prefill_handoff[tinyllama] - - disaggregated/test_ad_disagg.py::test_chunked_prefill_handoff[deepseek_v3_mla] -- condition: - ranges: - system_gpu_count: - gte: 1 - lte: 1 - wildcards: - gpu: - - '*h100*' - linux_distribution_name: ubuntu* - terms: - stage: post_merge - backend: autodeploy - orchestrator: mpi - tests: - - accuracy/test_llm_api_autodeploy.py::TestLlama3_1_8B::test_auto_dtype[trtllm-True-1] - - accuracy/test_llm_api_autodeploy.py::TestLlama3_1_8B_Instruct_Eagle3::test_eagle3_one_model[flashinfer] - - accuracy/test_llm_api_autodeploy.py::TestNemotronV2::test_auto_dtype[False] - - accuracy/test_llm_api_autodeploy.py::TestNemotronV2::test_auto_dtype[True] - - accuracy/test_llm_api_autodeploy.py::TestNemotronNanoV3::test_accuracy[bf16-1-attn_dp_off-trtllm] diff --git a/tests/integration/test_lists/test-db/l0_l40s.yml b/tests/integration/test_lists/test-db/l0_l40s.yml index 4deb7d3893c3..acaf44ff5818 100644 --- a/tests/integration/test_lists/test-db/l0_l40s.yml +++ b/tests/integration/test_lists/test-db/l0_l40s.yml @@ -29,9 +29,6 @@ l0_l40s: # Qwen3-Embedding text-embedding accuracy vs HuggingFace (0.6B small/fast + 8B large) - accuracy/test_llm_api_pytorch_encode.py::TestEncoderEncode::test_qwen3_text_embedding_matches_huggingface[qwen3-embedding-0.6b] - accuracy/test_llm_api_pytorch_encode.py::TestEncoderEncode::test_qwen3_text_embedding_matches_huggingface[qwen3-embedding-8b] - # AutoDeploy: Nemotron-Nano-V3 on Ada uses flashinfer; trtllm has no - # (E4M3 input, BF16 output, paged_kv, head_dim=128, sm_89) FMHA cubin. - - accuracy/test_llm_api_autodeploy.py::TestNemotronNanoV3::test_accuracy[fp8-1-attn_dp_off-flashinfer] # PyTorch: force NVFP4 Marlin on Ada to cover the SM89-compiled Marlin kernels. - accuracy/test_llm_api_pytorch.py::TestNemotronV3Nano::test_nvfp4_marlin_multi_gpus[tp_size=1] # ------------- Encoder-decoder tests --------------- diff --git a/tests/integration/test_lists/waives.txt b/tests/integration/test_lists/waives.txt index e6be6178e487..eb9c8981edbf 100644 --- a/tests/integration/test_lists/waives.txt +++ b/tests/integration/test_lists/waives.txt @@ -16,9 +16,6 @@ accuracy/test_disaggregated_serving.py::TestQwen3NextInstruct::test_auto_dtype[u accuracy/test_disaggregated_serving.py::TestQwen3_30B_A3B::test_mixed_ctx_gen_model[ctxpp2gentp2] SKIP (https://nvbugs/5748664) accuracy/test_dwdp_aggregated.py::TestDwdpAggDeepSeekV3Lite::test_dwdp_agg_accuracy[mode_a_uniform] SKIP (https://nvbugs/6661863) accuracy/test_dwdp_aggregated.py::TestDwdpAggDeepSeekV3Lite::test_dwdp_agg_accuracy[mode_a_uniform_contention_opt] SKIP (https://nvbugs/6644489) -accuracy/test_llm_api_autodeploy.py::TestNemotronNanoV3::test_accuracy[nvfp4-4-attn_dp_off-trtllm] SKIP (https://nvbugs/6367792) -accuracy/test_llm_api_autodeploy.py::TestNemotronSuperV3::test_mtp[nvfp4_ws8_80gb-trtllm] SKIP (https://nvbugs/6450341) -accuracy/test_llm_api_autodeploy.py::TestQwen3_5_397B_MoE::test_bf16_small[4] SKIP (https://nvbugs/6507114) accuracy/test_llm_api_pytorch.py::TestDeepSeekR1::test_fp8_blockscale[throughput_mtp] SKIP (https://nvbugs/6428101) accuracy/test_llm_api_pytorch.py::TestDeepSeekR1::test_fp8_blockscale[throughput_mtp_trtllm] SKIP (https://nvbugs/6426868) accuracy/test_llm_api_pytorch.py::TestDeepSeekR1::test_nvfp4_multi_gpus[latency_trtllmgen_adp_lmtp] SKIP (https://nvbugs/6655987) @@ -109,8 +106,6 @@ disaggregated/test_disaggregated.py::test_disaggregated_stress_test[input8k-outp disaggregated/test_workers.py::test_workers_conversation_router[TinyLlama-1.1B-Chat-v1.0] SKIP (https://nvbugs/6162322) disaggregated/test_workers.py::test_workers_kv_cache_aware_router_deepseek_v3_lite_bf16[DeepSeek-V3-Lite-bf16] SKIP (https://nvbugs/6162322) disaggregated/test_workers.py::test_workers_kv_cache_aware_router_eviction[TinyLlama-1.1B-Chat-v1.0] SKIP (https://nvbugs/6162322) -examples/test_ad_speculative_decoding.py::test_autodeploy_eagle3_one_model_acceptance_rate[trtllm-torch-cudagraph] SKIP (https://nvbugs/6426841) -examples/test_ad_speculative_decoding.py::test_nemotron_mtp_model_with_weights SKIP (https://nvbugs/6630699) examples/test_ray.py::test_ray_disaggregated_serving_python[tp2] SKIP (https://nvbugs/6601574) examples/visual_gen/test_visual_gen_cosmos3.py::test_cosmos3_feature_accuracy_against_golden[nvfp4] SKIP (https://nvbugs/6572800) examples/visual_gen/test_visual_gen_cosmos3.py::test_cosmos3_nano_t2v_lpips_against_golden SKIP (https://nvbugs/6655359) @@ -159,7 +154,6 @@ full:B200/llmapi/test_llm_api_pytorch_moe_lora.py::test_qwen_moe_routed_expert_m full:B200/test_e2e.py::test_multi_nodes_eval[Qwen3/Qwen3-235B-A22B-tp16-mmlu] SKIP (https://nvbugs/6424188) full:B200/test_e2e.py::test_multi_nodes_eval[Qwen3/saved_models_Qwen3-235B-A22B_nvfp4_hf-tp16-mmlu] SKIP (https://nvbugs/6424188) full:B300/accuracy/test_disaggregated_serving.py::TestDeepSeekV3Lite::test_auto_dtype_with_helix[fifo_v2-cudagraph:with_padding-pp1tp2cp2-overlap_off] SKIP (https://nvbugs/6410881) -full:B300/accuracy/test_llm_api_autodeploy.py::TestNemotronNanoV3::test_accuracy[bf16-4-attn_dp_off-trtllm] SKIP (https://nvbugs/6539942) full:B300/accuracy/test_llm_api_pytorch.py::TestDeepSeekV32::test_nvfp4_multi_gpus[latency] SKIP (https://nvbugs/6423866) full:B300/accuracy/test_llm_api_pytorch.py::TestDeepSeekV32::test_nvfp4_multi_gpus_chunked_prefill[latency_qsplit] SKIP (https://nvbugs/6423866) full:B300/accuracy/test_llm_api_pytorch.py::TestMiniMaxM3::test_auto_dtype[tp_size=8-ep_size=8] SKIP (https://nvbugs/6445375) @@ -183,7 +177,6 @@ full:GB200/accuracy/test_disaggregated_serving.py::TestLlama3_1_8BInstruct::test full:GB200/accuracy/test_dwdp_disaggregated_serving.py::TestDwdpDeepSeekV3Lite::test_dwdp_accuracy SKIP (https://nvbugs/6276923) full:GB200/accuracy/test_dwdp_disaggregated_serving.py::TestDwdpDeepSeekV3Lite::test_dwdp_accuracy_contention_opt SKIP (https://nvbugs/6276923) full:GB200/accuracy/test_dwdp_disaggregated_serving.py::TestDwdpDeepSeekV3Lite::test_dwdp_accuracy_mode_b_overlap SKIP (https://nvbugs/6276923) -full:GB200/accuracy/test_llm_api_autodeploy.py::TestNemotronNanoV3::test_accuracy[fp8-4-attn_dp_off-trtllm] SKIP (https://nvbugs/6539942) full:GB200/accuracy/test_llm_api_pytorch.py::TestDeepSeekV3Lite::test_bfloat16[mtp_nextn=0-attention_dp=False-cuda_graph=False-overlap_scheduler=False-torch_compile=False-enable_chunked_prefill=False-v2_kv_cache=False] SKIP (https://nvbugs/6525896) full:GB200/accuracy/test_llm_api_pytorch.py::TestDeepSeekV3Lite::test_cute_dsl_bf16_gemm[cuda_graph=True] SKIP (https://nvbugs/6525897) full:GB200/accuracy/test_llm_api_pytorch.py::TestMiniMaxM3::test_mxfp8[use_msa=False] SKIP (https://nvbugs/6479471) @@ -193,15 +186,8 @@ full:GB200/accuracy/test_llm_api_pytorch.py::TestQwen3_30B_A3B::test_nvfp4[dep4_ full:GB200/accuracy/test_llm_api_pytorch.py::TestQwen3_5_35B_A3B::test_fp8_moe_dflash SKIP (https://nvbugs/6316985) full:GB200/accuracy/test_llm_api_pytorch_multimodal.py::TestExaone4_5_33B::test_auto_dtype[forced_chunked_prefill] SKIP (https://nvbugs/6597570) full:GB200/accuracy/test_llm_api_pytorch_multimodal.py::TestExaone4_5_33B::test_auto_dtype[full_budget] SKIP (https://nvbugs/6597570) -full:GB200/disaggregated/test_ad_disagg.py::test_async_eagle3_full_model_handoff SKIP (https://nvbugs/6369254) -full:GB200/disaggregated/test_ad_disagg.py::test_async_generation_matches_aggregate SKIP (https://nvbugs/6415323) -full:GB200/disaggregated/test_ad_disagg.py::test_async_generation_no_overlap_matches_aggregate SKIP (https://nvbugs/6402495) -full:GB200/disaggregated/test_ad_disagg.py::test_async_sharded_generation_handoff SKIP (https://nvbugs/6402495) -full:GB200/disaggregated/test_ad_disagg_trtllm_serve.py::test_openai_completion SKIP (https://nvbugs/6402495) full:GB200/disaggregated/test_disaggregated.py::test_disaggregated_logprobs_serving[llama-3.1-8b-instruct] SKIP (https://nvbugs/6275959) full:GB300/accuracy/test_disaggregated_serving.py::TestDeepSeekV3Lite::test_gen_first[adp-mtp2] SKIP (https://nvbugs/6295740) -full:GB300/accuracy/test_llm_api_autodeploy.py::TestNemotronNanoV3::test_accuracy[bf16-4-attn_dp_off-trtllm] SKIP (https://nvbugs/6539942) -full:GB300/accuracy/test_llm_api_autodeploy.py::TestNemotronNanoV3::test_accuracy[nvfp4-1-attn_dp_off-trtllm] SKIP (https://nvbugs/6329165) full:GB300/accuracy/test_llm_api_pytorch.py::TestDeepSeekV3Lite::test_bfloat16_4gpus[pp4-mtp_nextn=2-attention_dp=True-cuda_graph=True-overlap_scheduler=True-torch_compile=True] SKIP (https://nvbugs/6432818) full:GB300/accuracy/test_llm_api_pytorch.py::TestDeepSeekV3Lite::test_fp8_block_scales_4gpus[pp4-mtp_nextn=0-fp8kv=False-attention_dp=True-cuda_graph=True-overlap_scheduler=True-torch_compile=False-sampler_async_worker=False] SKIP (https://nvbugs/6661948) full:GB300/accuracy/test_llm_api_pytorch.py::TestDeepSeekV3Lite::test_fp8_block_scales_4gpus[pp4-mtp_nextn=0-fp8kv=True-attention_dp=False-cuda_graph=True-overlap_scheduler=True-torch_compile=True-sampler_async_worker=False] SKIP (https://nvbugs/6661948) @@ -374,12 +360,6 @@ unittest/_torch/visual_gen/multi_gpu/test_wan_pipeline_parallel.py::TestWanPipel unittest/_torch/visual_gen/test_wan22_i2v_pipeline.py SKIP (https://nvbugs/6626445) unittest/_torch/visual_gen/test_wan22_ti2v_5b_pipeline.py::TestWan22TI2V5BCombinedOptimizations::test_fp8_cache_dit_trtllm SKIP (https://nvbugs/6661914) unittest/_torch/visual_gen/test_wan22_ti2v_5b_pipeline.py::TestWan22TI2V5B_T2V_PipelineCorrectness::test_cosine_similarity SKIP (https://nvbugs/6661914) -unittest/auto_deploy/multigpu/custom_ops SKIP (https://nvbugs/6403920) -unittest/auto_deploy/singlegpu/models/test_glm4_moe_modeling.py::test_glm4_moe_decoder_layer_equivalence[1-dtype0-1-8] SKIP (https://nvbugs/6566730) -unittest/auto_deploy/singlegpu/models/test_glm4_moe_modeling.py::test_glm4_moe_moe_equivalence[dtype0-1-8] SKIP (https://nvbugs/6566730) -unittest/auto_deploy/singlegpu/shim/test_ad_executor_swa_eviction.py SKIP (https://nvbugs/6606123) -unittest/auto_deploy/singlegpu/shim/test_cached_sequence_interface.py::test_initialize_resources_mixed_creates_mamba_hybrid_cache_manager SKIP (https://nvbugs/6606123) -unittest/auto_deploy/standalone/test_standalone_package.py::TestStandalonePackage::test_run_unit_tests SKIP (https://nvbugs/6670106) unittest/bindings/test_transfer_agent_bindings.py::TestMooncakeFunctionalTransfer::test_mooncake_wait_in_progress_on_zero_timeout SKIP (https://nvbugs/6517836) unittest/bindings/test_transfer_agent_bindings.py::TestMooncakeFunctionalTransfer::test_mooncake_write_transfer_gpu_tensor SKIP (https://nvbugs/6517836) unittest/bindings/test_transfer_agent_bindings.py::TestMooncakeFunctionalTransfer::test_mooncake_write_transfer_multiple_chunks SKIP (https://nvbugs/6517836)