Merge tag 'v0.7.2' into v0.7.2-dev

vllm/spec_decode/medusa_worker.py

+# SPDX-License-Identifier: Apache-2.0
+
 import os
 import weakref
 from typing import List, Optional, Set, Tuple, Dict

vllm/spec_decode/metrics.py

+# SPDX-License-Identifier: Apache-2.0
+
 import time
 from typing import Callable, Optional, Union
 

vllm/spec_decode/mlp_speculator_worker.py

+# SPDX-License-Identifier: Apache-2.0
+
 from typing import List, Optional, Set, Tuple, Dict
 
 import torch

vllm/spec_decode/mqa_scorer.py

+# SPDX-License-Identifier: Apache-2.0
+
 from vllm.sequence import (ExecuteModelRequest, SequenceData,
                            SequenceGroupMetadata, get_all_seq_ids)
 from vllm.spec_decode.interfaces import (SpeculativeProposals,

vllm/spec_decode/multi_step_worker.py

+# SPDX-License-Identifier: Apache-2.0
+
 import copy
 import weakref
 from typing import Dict, List, Set, Tuple

vllm/spec_decode/ngram_worker.py

+# SPDX-License-Identifier: Apache-2.0
+
 import weakref
 from typing import List, Optional, Set, Tuple
 

vllm/spec_decode/proposer_worker_base.py

+# SPDX-License-Identifier: Apache-2.0
+
 from abc import ABC, abstractmethod
 from typing import List, Optional, Set, Tuple
 

vllm/spec_decode/smaller_tp_proposer_worker.py

+# SPDX-License-Identifier: Apache-2.0
+
 from typing import List, Optional, Set, Tuple
 
 import torch

vllm/spec_decode/spec_decode_worker.py

+# SPDX-License-Identifier: Apache-2.0
+
 import os
 import copy
 from collections import defaultdict

vllm/spec_decode/target_model_runner.py

+# SPDX-License-Identifier: Apache-2.0
+
 from typing import List, Optional
 
 from vllm.sequence import SequenceGroupMetadata

vllm/spec_decode/top1_proposer.py

+# SPDX-License-Identifier: Apache-2.0
+
 from typing import List, Optional, Set, Tuple
 
 import torch

vllm/spec_decode/util.py

+# SPDX-License-Identifier: Apache-2.0
+
 import time
 from contextlib import contextmanager
 from typing import Dict, List, Optional, Sequence, Tuple

vllm/tracing.py

+# SPDX-License-Identifier: Apache-2.0
+
 import os
 from typing import Mapping, Optional
 

vllm/transformers_utils/__init__.py

+# SPDX-License-Identifier: Apache-2.0
+
 from vllm.envs import VLLM_USE_MODELSCOPE
 
 if VLLM_USE_MODELSCOPE:

vllm/transformers_utils/config.py

+# SPDX-License-Identifier: Apache-2.0
+
 import enum
 import json
 import os

vllm/transformers_utils/configs/__init__.py

+# SPDX-License-Identifier: Apache-2.0
+
 from vllm.transformers_utils.configs.chatglm import ChatGLMConfig
 from vllm.transformers_utils.configs.cohere2 import Cohere2Config
 from vllm.transformers_utils.configs.dbrx import DbrxConfig

vllm/transformers_utils/configs/arctic.py

+# SPDX-License-Identifier: Apache-2.0
+
 # yapf: disable
 # ruff: noqa: E501
 # coding=utf-8

vllm/transformers_utils/configs/chatglm.py

+# SPDX-License-Identifier: Apache-2.0
+
 # Adapted from
 # https://github.com/THUDM/ChatGLM2-6B
 from transformers import PretrainedConfig

vllm/transformers_utils/configs/cohere2.py

+# SPDX-License-Identifier: Apache-2.0
+
 # ruff: noqa
 
 # Adapted from

vllm/transformers_utils/configs/dbrx.py

+# SPDX-License-Identifier: Apache-2.0
+
 # yapf: disable
 # ruff: noqa: E501
 # coding=utf-8