diff --git a/tests/integration/test_lists/waives.txt b/tests/integration/test_lists/waives.txt index 659836db32d8..a0498595087f 100644 --- a/tests/integration/test_lists/waives.txt +++ b/tests/integration/test_lists/waives.txt @@ -358,7 +358,6 @@ unittest/_torch/thop/serial/test_moe.py::TestMoeFp4::test_no_autotune[use_score_ unittest/_torch/thop/serial/test_moe.py::TestMoeFp4::test_no_autotune[use_score_as_input-RoutingRenormalize_qwen_next-swiglu-1024-1024-150] SKIP (https://nvbugs/5908070) unittest/_torch/thop/serial/test_moe.py::TestMoeFp4::test_no_autotune[use_score_as_input-RoutingRenormalize_topk_4-swiglu-1024-1024-150] SKIP (https://nvbugs/5908070) unittest/_torch/visual_gen/test_flux_pipeline.py::TestFluxCombinedOptimizations::test_all_optimizations_combined SKIP (https://nvbugs/6199854) -unittest/auto_deploy/singlegpu/shim/test_llm_config.py::test_speculative_flashinfer_fallback_disables_piecewise SKIP (https://nvbugs/6309375) unittest/auto_deploy/singlegpu/smoke SKIP (https://nvbugs/6306936) unittest/auto_deploy/singlegpu/smoke/test_disagg.py::test_autodeploy_disaggregated_batch_smoke[deepseek-trtllm-simple] SKIP (https://nvbugs/6307525) unittest/auto_deploy/singlegpu/smoke/test_disagg.py::test_autodeploy_disaggregated_batch_smoke[llama-flashinfer-cudagraph] SKIP (https://nvbugs/6307525) diff --git a/tests/unittest/auto_deploy/singlegpu/shim/test_llm_config.py b/tests/unittest/auto_deploy/singlegpu/shim/test_llm_config.py index 1c3212509354..85553ebab42a 100644 --- a/tests/unittest/auto_deploy/singlegpu/shim/test_llm_config.py +++ b/tests/unittest/auto_deploy/singlegpu/shim/test_llm_config.py @@ -95,25 +95,6 @@ def test_yaml_compile_backend_disables_default_piecewise(tmp_path): assert args.transforms["compile_model"]["piecewise_enabled"] is False -def test_speculative_flashinfer_fallback_disables_piecewise(): - from tensorrt_llm.llmapi import EagleDecodingConfig - - spec_config = EagleDecodingConfig( - max_draft_len=3, - speculative_model="some/model", - eagle3_one_model=True, - ) - - args = LlmArgs( - model="test-model", - attn_backend="flashinfer", - speculative_config=spec_config, - ) - - assert args.compile_backend == "torch-simple" - assert args.transforms["compile_model"]["piecewise_enabled"] is False - - def test_cache_transceiver_rejects_unmanaged_persistent_caches(): """Cache transceiver rejects unmanaged persistent cache resources.""" args = LlmArgs(