Merge branch 'main' into refactorTRTLLMGen

Author: xxi-nv

tensorrt_llm/_torch/pyexecutor/model_engine.py

@@ -181,7 +181,11 @@ def __init__(
 
         self.attn_runtime_features = attn_runtime_features or AttentionRuntimeFeatures(
         )
-        self.input_processor = create_input_processor(model_path, None)
+
+        self.input_processor = create_input_processor(
+            model_path,
+            tokenizer=None,
+            checkpoint_format=llm_args.checkpoint_format)
         self.input_processor_with_hash = create_input_processor_with_hash(
             self.input_processor)
         if model is None:

tests/integration/test_lists/waives.txt

@@ -401,3 +401,15 @@ accuracy/test_disaggregated_serving.py::TestLlama3_1_8BInstruct::test_auto_dtype
 accuracy/test_llm_api_autodeploy.py::TestNemotronH::test_auto_dtype[True] SKIP (https://nvbugs/5688721)
 unittest/_torch/speculative/test_eagle3.py::test_llama_eagle3[True-FLASHINFER-False-False-False-False-True-False-False] SKIP (https://nvbugs/5691246)
 accuracy/test_llm_api_pytorch.py::TestDeepSeekR1::test_nvfp4_multi_gpus[throughput_mtp] SKIP (https://nvbugs/5698897)
+test_e2e.py::test_openai_completions_example[trt] SKIP (https://nvbugs/5701450)
+examples/test_ray.py::test_llm_inference_distributed_ray[tep2] SKIP (https://nvbugs/5701457)
+accuracy/test_llm_api_pytorch.py::TestLlama3_1_8BInstruct::test_fp8_4gpus[tp4-fp8kv=False-attn_backend=TRTLLM-torch_compile=False] SKIP (https://nvbugs/5701457)
+accuracy/test_llm_api_pytorch.py::TestDeepSeekV3Lite::test_fp8_block_scales_4gpus[tp4-mtp_nextn=2-fp8kv=True-attention_dp=False-cuda_graph=True-overlap_scheduler=True-torch_compile=True] SKIP (https://nvbugs/5701445)
+accuracy/test_llm_api_pytorch.py::TestDeepSeekV3Lite::test_nvfp4_4gpus[moe_backend=CUTLASS-mtp_nextn=0-tp2pp2-fp8kv=False-attention_dp=False-cuda_graph=False-overlap_scheduler=False-torch_compile=False] SKIP (https://nvbugs/5666821)
+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] SKIP (https://nvbugs/5666821)
+accuracy/test_llm_api_pytorch.py::TestDeepSeekV3Lite::test_nvfp4_4gpus[moe_backend=CUTLASS-mtp_nextn=0-ep4-fp8kv=False-attention_dp=False-cuda_graph=False-overlap_scheduler=False-torch_compile=False] SKIP (https://nvbugs/5666821)
+triton_server/test_triton_llm.py::test_llmapi_backend[4-0-disableDecoupleMode-tensorrt_llm] SKIP (https://nvbugs/5701480)
+accuracy/test_disaggregated_serving.py::TestDeepSeekV3Lite::test_auto_dtype[mtp_nextn=0-overlap_scheduler=False] SKIP (https://nvbugs/5701491)
+accuracy/test_llm_api_pytorch.py::TestDeepSeekV3Lite::test_bfloat16_4gpus[ep4-mtp_nextn=0-attention_dp=False-cuda_graph=False-overlap_scheduler=False-torch_compile=False] SKIP (https://nvbugs/5701425)
+accuracy/test_llm_api_pytorch.py::TestDeepSeekV3Lite::test_bfloat16_4gpus[tp4-mtp_nextn=2-attention_dp=False-cuda_graph=True-overlap_scheduler=True-torch_compile=False] SKIP (https://nvbugs/5701425)
+accuracy/test_llm_api_pytorch.py::TestDeepSeekR1::test_nvfp4_multi_gpus[throughput_tp8] SKIP (https://nvbugs/5698897)