[Core][Performance] Add XGrammar support for guided decoding and set it as default #28123
Annotations
10 errors
Analysing the code with ruff:
vllm/config.py#L1998
vllm/config.py:1998:81: E501 Line too long (87 > 80)
|
Analysing the code with ruff:
vllm/model_executor/guided_decoding/__init__.py#L3
vllm/model_executor/guided_decoding/__init__.py:3:30: F401 `typing.Any` imported but unused
|
Analysing the code with ruff:
vllm/model_executor/guided_decoding/__init__.py#L14
vllm/model_executor/guided_decoding/__init__.py:14:39: UP007 Use `X | Y` for type annotations
|
Analysing the code with ruff:
vllm/model_executor/guided_decoding/__init__.py#L28
vllm/model_executor/guided_decoding/__init__.py:28:81: E501 Line too long (125 > 80)
|
Analysing the code with ruff:
vllm/model_executor/guided_decoding/__init__.py#L39
vllm/model_executor/guided_decoding/__init__.py:39:39: UP007 Use `X | Y` for type annotations
|
Analysing the code with ruff:
vllm/model_executor/guided_decoding/__init__.py#L53
vllm/model_executor/guided_decoding/__init__.py:53:81: E501 Line too long (125 > 80)
|
Analysing the code with ruff:
vllm/model_executor/guided_decoding/xgrammar_decoding.py#L3
vllm/model_executor/guided_decoding/xgrammar_decoding.py:3:1: E401 Multiple imports on one line
|
Analysing the code with ruff:
vllm/model_executor/guided_decoding/xgrammar_decoding.py#L28
vllm/model_executor/guided_decoding/xgrammar_decoding.py:28:81: E501 Line too long (109 > 80)
|
Analysing the code with ruff:
vllm/model_executor/guided_decoding/xgrammar_decoding.py#L36
vllm/model_executor/guided_decoding/xgrammar_decoding.py:36:15: UP007 Use `X | Y` for type annotations
|
Analysing the code with ruff:
vllm/model_executor/guided_decoding/xgrammar_decoding.py#L37
vllm/model_executor/guided_decoding/xgrammar_decoding.py:37:18: UP007 Use `X | Y` for type annotations
|
Loading