Merge tag 'v0.9.1' into v0.9.1-ori

tests/quantization/test_experts_int8.py

 # SPDX-License-Identifier: Apache-2.0
+# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
 
 # flake8: noqa
 """Tests experts_int8 quantization startup and generation, 

tests/quantization/test_fp8.py

 # SPDX-License-Identifier: Apache-2.0
+# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
 """Tests whether FP8 computation is enabled correctly.
 
 Run `pytest tests/quantization/test_fp8.py --forked`.

tests/quantization/test_gptq_dynamic.py

 # SPDX-License-Identifier: Apache-2.0
+# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
 """Tests whether gptq models with dynamic quantized can be loaded.
 
 Run `pytest tests/quantization/test_gptq_dynamic.py --forked`.

tests/quantization/test_ipex_quant.py

 # SPDX-License-Identifier: Apache-2.0
+# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
 """Test model set-up and inference for quantized HF models supported
  on the CPU/GPU backend using IPEX (including AWQ/GPTQ).
  

tests/quantization/test_lm_head.py

 # SPDX-License-Identifier: Apache-2.0
+# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
 """Tests whether gptq models with quantized lm_head can be loaded.
 
 Run `pytest tests/quantization/test_quant_lm_head_true.py --forked`.

tests/quantization/test_ptpc_fp8.py

 # SPDX-License-Identifier: Apache-2.0
+# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
 """Tests whether PTPC w8a8 FP8 computation is enabled correctly.
 
 Run `pytest tests/quantization/test_ptpc_fp8.py --forked`.

tests/quantization/test_quark.py

 # SPDX-License-Identifier: Apache-2.0
+# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
 """Test model set-up and weight loading for quark-quantized models.
 
 Run `pytest tests/quantization/test_quark.py`.

tests/quantization/test_register_quantization_config.py

 # SPDX-License-Identifier: Apache-2.0
+# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
 """Tests register custom quantization config.
 
 See https://github.com/vllm-project/vllm/issues/11926 for more details.

tests/quantization/test_torchao.py

 # SPDX-License-Identifier: Apache-2.0
+# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
 import importlib.metadata
 import importlib.util
 
@@ -12,7 +13,7 @@ TORCHAO_AVAILABLE = importlib.util.find_spec("torchao") is not None
 
 @pytest.mark.skipif(not TORCHAO_AVAILABLE, reason="torchao is not available")
 def test_pre_quantized_model(vllm_runner):
-    with vllm_runner("drisspg/float8_dynamic_act_float8_weight-opt-125m",
+    with vllm_runner("drisspg/fp8-opt-125m",
                      quantization="torchao",
                      dtype="bfloat16",
                      enforce_eager=True) as llm:
@@ -29,10 +30,10 @@ def test_pre_quantized_model(vllm_runner):
         "cuda:0",
         # {"": "cuda"},
     ])
-def test_opt_125m_int4wo_model_loading_with_params(vllm_runner,
+def test_opt_125m_int8wo_model_loading_with_params(vllm_runner,
                                                    pt_load_map_location):
     torch._dynamo.reset()
-    model_name = "jerryzh168/opt-125m-int4wo"
+    model_name = "jerryzh168/opt-125m-int8wo-partial-quant"
     with vllm_runner(model_name=model_name,
                      quantization="torchao",
                      dtype="bfloat16",

tests/quantization/utils.py

 # SPDX-License-Identifier: Apache-2.0
+# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
 
 from vllm.model_executor.layers.quantization import get_quantization_config
 from vllm.platforms import current_platform

tests/reasoning/test_deepseekr1_reasoning_parser.py

 # SPDX-License-Identifier: Apache-2.0
+# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
 
 import pytest
 from transformers import AutoTokenizer

tests/reasoning/test_granite_reasoning_parser.py

 # SPDX-License-Identifier: Apache-2.0
+# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
 import pytest
 from transformers import AutoTokenizer
 

tests/reasoning/test_qwen3_reasoning_parser.py

 # SPDX-License-Identifier: Apache-2.0
+# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
 
 import pytest
 from transformers import AutoTokenizer

tests/reasoning/utils.py

 # SPDX-License-Identifier: Apache-2.0
+# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
 
 from typing import Optional, Union
 

tests/runai_model_streamer_test/test_runai_model_streamer_loader.py

 # SPDX-License-Identifier: Apache-2.0
+# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
 
 from vllm import SamplingParams
 from vllm.config import LoadConfig, LoadFormat

tests/runai_model_streamer_test/test_weight_utils.py

 # SPDX-License-Identifier: Apache-2.0
+# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
 
 import glob
 import tempfile

tests/samplers/test_beam_search.py

 # SPDX-License-Identifier: Apache-2.0
+# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
 """Compare the outputs of HF and vLLM when using beam search.
 
 Run `pytest tests/samplers/test_beam_search.py`.

tests/samplers/test_ignore_eos.py

 # SPDX-License-Identifier: Apache-2.0
+# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
 """Make sure ignore_eos works.
 
 Run `pytest tests/samplers/test_ignore_eos.py`.

tests/samplers/test_logits_processor.py

 # SPDX-License-Identifier: Apache-2.0
+# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
 
 import pytest
 import torch

tests/samplers/test_logprobs.py

 # SPDX-License-Identifier: Apache-2.0
+# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
 
 import pytest
 import torch