vllm.LLM 的参数

版本

vllm 0.7.3

源码

参考链接：https://github.com/vllm-project/vllm/blob/36e0c8f7da96b69abf7e8b19527c0ee953805395/vllm/entrypoints/llm.py#L53

class LLM:
    """An LLM for generating texts from given prompts and sampling parameters.

    This class includes a tokenizer, a language model (possibly distributed
    across multiple GPUs), and GPU memory space allocated for intermediate
    states (aka KV cache). Given a batch of prompts and sampling parameters,
    this class generates texts from the model, using an intelligent batching
    mechanism and efficient memory management.

    Args:
        model: The name or path of a HuggingFace Transformers model.
        tokenizer: The name or path of a HuggingFace Transformers tokenizer.
        tokenizer_mode: The tokenizer mode. "auto" will use the fast tokenizer
            if available, and "slow" will always use the slow tokenizer.
        skip_tokenizer_init: If true, skip initialization of tokenizer and
            detokenizer. Expect valid prompt_token_ids and None for prompt
            from the input.
        trust_remote_code: Trust remote code (e.g., from HuggingFace) when
            downloading the model and tokenizer.
        allowed_local_media_path: Allowing API requests to read local images
            or videos from directories specified by the server file system.
            This is a security risk. Should only be enabled in trusted
            environments.
        tensor_parallel_size: The number of GPUs to use for distributed
            execution with tensor parallelism.
        dtype: The data type for the model weights and activations. Currently,
            we support `float32`, `float16`, and `bfloat16`. If `auto`, we use
            the `torch_dtype` attribute specified in the model config file.
            However, if the `torch_dtype` in the config is `float32`, we will
            use `float16` instead.
        quantization: The method used to quantize the model weights. Currently,
            we support "awq", "gptq", and "fp8" (experimental).
            If None, we first check the `quantization_config` attribute in the
            model config file. If that is None, we assume the model weights are
            not quantized and use `dtype` to determine the data type of
            the weights.
        revision: The specific model version to use. It can be a branch name,
            a tag name, or a commit id.
        tokenizer_revision: The specific tokenizer version to use. It can be a
            branch name, a tag name, or a commit id.
        seed: The seed to initialize the random number generator for sampling.
        gpu_memory_utilization: The ratio (between 0 and 1) of GPU memory to
            reserve for the model weights, activations, and KV cache. Higher
            values will increase the KV cache size and thus improve the model's
            throughput. However, if the value is too high, it may cause out-of-
            memory (OOM) errors.
        swap_space: The size (GiB) of CPU memory per GPU to use as swap space.
            This can be used for temporarily storing the states of the requests
            when their `best_of` sampling parameters are larger than 1. If all
            requests will have `best_of=1`, you can safely set this to 0.
            Noting that `best_of` is only supported in V0. Otherwise, too small
            values may cause out-of-memory (OOM) errors.
        cpu_offload_gb: The size (GiB) of CPU memory to use for offloading
            the model weights. This virtually increases the GPU memory space
            you can use to hold the model weights, at the cost of CPU-GPU data
            transfer for every forward pass.
        enforce_eager: Whether to enforce eager execution. If True, we will
            disable CUDA graph and always execute the model in eager mode.
            If False, we will use CUDA graph and eager execution in hybrid.
        max_seq_len_to_capture: Maximum sequence len covered by CUDA graphs.
            When a sequence has context length larger than this, we fall back
            to eager mode. Additionally for encoder-decoder models, if the
            sequence length of the encoder input is larger than this, we fall
            back to the eager mode.
        disable_custom_all_reduce: See :class:`~vllm.config.ParallelConfig`
        disable_async_output_proc: Disable async output processing.
            This may result in lower performance.
        hf_overrides: If a dictionary, contains arguments to be forwarded to the
            HuggingFace config. If a callable, it is called to update the
            HuggingFace config.
        compilation_config: Either an integer or a dictionary. If it is an
            integer, it is used as the level of compilation optimization. If it
            is a dictionary, it can specify the full compilation configuration.
        **kwargs: Arguments for :class:`~vllm.EngineArgs`. (See
            :ref:`engine-args`)

    Note:
        This class is intended to be used for offline inference. For online
        serving, use the :class:`~vllm.AsyncLLMEngine` class instead.
    """

参数解释

---

vllm.LLM 类是 vLLM 库的核心，用于加载和管理大型语言模型，并进行高效的文本生成。它包含了模型、分词器以及为中间状态（KV 缓存）分配的 GPU 内存空间。

以下是 vllm.LLM 构造函数中各个参数的详细解释：

1. model (str)

• 类型: 字符串
• 描述: 指定要加载的 HuggingFace Transformers 模型的名称或本地路径。
• 取值:
  • Hugging Face 模型 Hub 名称: 例如 "facebook/opt-350m", "EleutherAI/gpt-neo-2.7B", "meta-llama/Llama-2-7b-hf", "mistralai/Mistral-7B-v0.1" 等。vLLM 会自动从 Hugging Face Model Hub 下载模型配置和权重。
  • 本地模型路径: 指向包含模型权重和配置文件的本地目录的路径。
• 作用: 这是最关键的参数，它定义了你想要使用的预训练语言模型。vLLM 会根据这个参数加载相应的模型架构和预训练权重。

2. tokenizer (str, 可选)

• 类型: 字符串，可选
• 描述: 指定要使用的 HuggingFace Transformers 分词器的名称或本地路径。
• 取值:
  • Hugging Face Tokenizer Hub 名称: 例如 "facebook/opt-350m", "EleutherAI/gpt-neo-2.7B", "meta-llama/Llama-2-7b-hf", "mistralai/Mistral-7B-v0.1" 等。 通常应与 model 参数指定的模型匹配。
  • 本地分词器路径: 指向包含分词器配置和文件的本地目录。
  • None (默认): 如果为 None，vLLM 会尝试从 model 参数指定的模型配置中自动推断并加载相应的分词器。
• 作用: 分词器负责将输入的文本转换为模型可以理解的 token ID 序列，以及将模型生成的 token ID 序列转换回可读的文本。 绝大多数情况下，让 vLLM 自动推断是最佳选择。只有在需要使用特别定制的分词器时才需要手动指定。

3. tokenizer_mode (str, 可选)

• 类型: 字符串，可选
• 描述: 分词器模式。
• 取值:
  • "auto" (默认): 如果可用，使用快速分词器（基于 Rust 实现），否则回退到慢速分词器（基于 Python 实现）。
  • "slow": 始终使用慢速分词器。
• 作用: 快速分词器通常性能更高，尤其是在批量处理时。 auto 模式是推荐的，它会在性能和兼容性之间做出权衡。

4. skip_tokenizer_init (bool, 可选)

• 类型: 布尔值，可选
• 描述: 如果为 True，则跳过分词器和反分词器的初始化。
• 取值:
  • True: 跳过初始化。
  • False (默认): 进行初始化。
• 作用: 当设置为 True 时，你需要在输入中直接提供 prompt_token_ids，并且 prompt 参数应为 None。 这适用于一些高级用例，例如你已经预先完成了分词，或者使用了自定义的分词流程。 通常情况下，应该保持默认值 False，让 vLLM 自动处理分词。

5. trust_remote_code (bool, 可选)

• 类型: 布尔值，可选
• 描述: 是否信任来自 HuggingFace 的远程代码（例如在下载模型和分词器时）。
• 取值:
  • True: 信任远程代码。
  • False (默认): 不信任远程代码。
• 作用: 某些模型或分词器可能包含需要执行的自定义代码（例如定义了自定义的模型架构或分词逻辑）。 如果设置为 True，vLLM 会执行这些远程代码。 安全性重要提示： 启用此选项存在安全风险，因为你信任了代码的提供者。 仅在可信环境下使用，并且当你确信需要运行远程代码时才应设置为 True。

6. allowed_local_media_path (str, 可选)

• 类型: 字符串，可选
• 描述: 允许 API 请求从服务器文件系统指定的目录读取本地图像或视频。
• 取值: 本地目录路径。
• 作用: 为了安全起见，默认情况下 API 请求无法直接访问本地文件。设置此参数可以允许 API 读取指定路径下的本地媒体文件，用于多模态模型的输入处理。 安全性重要提示: 这是一个安全风险，仅应在受信任的环境中启用。

7. tensor_parallel_size (int, 可选)

• 类型: 整数，可选
• 描述: 使用张量并行进行分布式执行的 GPU 数量。
• 取值: 正整数。 例如 1, 2, 4, 8 等。
• 作用: 张量并行是一种将模型权重和计算负载分布到多个 GPU 上以进行并行处理的技术。增加 tensor_parallel_size 可以利用多 GPU 的计算能力，加速推理并减少单个 GPU 的内存压力。 如果你的系统有多个 GPU，并且你想加载大型模型或提高吞吐量，应该考虑增加此参数。

8. dtype (str, 可选)

• 类型: 字符串，可选
• 描述: 模型权重和激活的数据类型。
• 取值:
  • "float32": 单精度浮点数。
  • "float16": 半精度浮点数。
  • "bfloat16": Brain floating-point 16 位浮点数。
  • "auto" (默认): 自动选择。 如果模型配置文件 (config.json) 中指定了 torch_dtype 属性，则使用该属性的值。 特殊情况： 如果配置文件中 torch_dtype 为 float32，vLLM 也会默认使用 float16 以提高性能并减少内存占用。
• 作用: 数据类型影响模型的精度、计算速度和内存占用。较低精度的数据类型（如 float16 和 bfloat16）可以减少内存使用并加速计算，但可能会略微降低模型精度。 auto 模式通常是推荐的，它会在性能和精度之间做出平衡。

9. quantization (str, 可选)

• 类型: 字符串，可选
• 描述: 模型权重所使用的量化方法。
• 取值:
  • "awq": 激活感知权重量化 (AWQ)。
  • "gptq": GPTQ 量化。
  • "fp8": FP8 量化 (实验性)。
  • None (默认): 不进行量化。 如果模型配置文件中存在 quantization_config 属性，则检查该属性。如果 quantization_config 也为 None，则假定模型权重未量化，并使用 dtype 参数指定的数据类型。
• 作用: 量化是一种减小模型大小、加速推理的技术。 AWQ、GPTQ 和 FP8 是几种流行的后训练量化方法。 量化可以在一定程度上压缩模型，并提升推理速度，尤其是在 GPU 内存受限的情况下。 选择合适的量化方法可以在模型大小、速度和精度之间找到平衡。

10. revision (str, 可选)

• 类型: 字符串，可选
• 描述: 要使用的特定模型版本。
• 取值: 分支名称、标签名称或 commit ID。
• 作用: 允许你精确地指定要加载的模型仓库的特定版本。 这对于复现实验结果或使用特定版本的模型至关重要。

11. tokenizer_revision (str, 可选)

• 类型: 字符串，可选
• 描述: 要使用的特定分词器版本。
• 取值: 分支名称、标签名称或 commit ID。
• 作用: 与 revision 类似，但针对分词器。 允许你精确地指定要加载的分词器仓库的特定版本.

12. seed (int, 可选)

• 类型: 整数，可选
• 描述: 用于初始化采样随机数生成器的种子。
• 取值: 整数。
• 作用: 设置随机种子可以使模型的采样过程具有可复现性。 对于相同的输入和相同的种子，模型输出的结果将是相同的。这对于调试、测试和实验复现非常重要。

13. gpu_memory_utilization (float, 可选)

• 类型: 浮点数，可选
• 描述: GPU 内存使用率，表示为 0 到 1 之间的比例。
• 取值: 0.0 到 1.0 之间的浮点数。
• 作用: 指定为模型权重、激活和 KV 缓存预留的 GPU 内存比例。 较高的值会增加 KV 缓存的大小，从而提高模型的吞吐量。 然而，如果值过高，可能会导致内存溢出 (OOM) 错误。 合理设置 gpu_memory_utilization 可以在性能和稳定性之间取得平衡。 当遇到 “KV cache space is not enough” 错误时，可以尝试 谨慎地增加 此值。

14. swap_space (int, 可选)

• 类型: 整数，可选
• 描述: 每个 GPU 使用的 CPU 内存作为交换空间的大小 (GiB)。
• 取值: 非负整数 (单位 GiB)。
• 作用: 当 GPU 内存不足以容纳所有请求的状态（例如 best_of 参数大于 1 的请求）时，可以将部分状态临时存储到 CPU 内存中（换出）。 如果所有请求的 best_of 参数都为 1，可以安全地设置为 0。 注意：best_of 参数仅在 vLLM V0 版本中支持。 设置过小的 swap_space 可能会导致 OOM 错误。 启用交换空间会降低性能，因为它涉及到 CPU 和 GPU 之间的数据传输。

15. cpu_offload_gb (int, 可选)

• 类型: 整数，可选
• 描述: 用于卸载模型权重的 CPU 内存大小 (GiB)。
• 取值: 非负整数 (单位 GiB)。
• 作用: 通过将部分模型权重卸载到 CPU 内存，虚拟地增加可以用于模型权重的 GPU 内存空间。 但每次前向传播都需要进行 CPU-GPU 数据传输，会降低推理速度。 这是一种在极度 GPU 内存受限的情况下加载大型模型的方法，需要权衡性能损失。

16. enforce_eager (bool, 可选)

• 类型: 布尔值，可选
• 描述: 是否强制执行 eager execution（eager 模式）。
• 取值:
  • True: 强制 eager execution。禁用 CUDA Graph，始终以 eager 模式执行模型。
  • False (默认): 混合使用 CUDA Graph 和 eager execution。 对于可以捕获到 CUDA Graph 中的计算图，使用 CUDA Graph 执行以加速；对于超出 CUDA Graph 捕获范围的计算，回退到 eager 模式。
• 作用: CUDA Graph 是一种优化技术，可以减少 GPU kernel 启动的开销，从而提高推理性能。 默认情况下 vLLM 会混合使用 CUDA Graph 和 eager execution 以获得最佳性能。 在某些调试或特定场景下，可能需要强制使用 eager execution。

17. max_seq_len_to_capture (int, 可选)

• 类型: 整数，可选
• 描述: CUDA Graph 可以覆盖的最大序列长度。
• 取值: 正整数。
• 作用: 当序列的上下文长度或 encoder-decoder 模型中 encoder 输入的序列长度大于此值时，会回退到 eager 模式执行。 设置此参数可以控制 CUDA Graph 编译的范围，避免编译过大的 CUDA Graph 导致内存占用过高或编译时间过长。

18. disable_custom_all_reduce (bool, 可选)

• 类型: 布尔值，可选
• 描述: 禁用自定义的 all_reduce 操作。
• 作用: 详细信息请参考 vllm.config.ParallelConfig 类文档。 这通常是一个底层配置参数，用于控制分布式训练或推理中的通信方式。 在大多数情况下，用户无需手动设置。

19. disable_async_output_proc (bool, 可选)

• 类型: 布尔值，可选
• 描述: 禁用异步输出处理。
• 取值:
  • True: 禁用异步输出处理。
  • False (默认): 启用异步输出处理。
• 作用: 异步输出处理可以提高吞吐量，但可能会导致输出结果的返回顺序与请求的顺序略有不同。 禁用异步输出处理可能会降低性能，但可以保证输出结果的顺序与请求顺序一致。

20. hf_overrides (dict or callable, 可选)

• 类型: 字典或可调用对象，可选
• 描述: 用于覆盖 HuggingFace 配置的参数。
• 取值:
  • 字典: 包含要传递给 HuggingFace config.from_pretrained 的参数。
  • 可调用对象: 一个函数，用于更新 HuggingFace 配置对象。
• 作用: 允许用户在 vLLM 加载模型时，对 HuggingFace 模型配置进行自定义修改。 例如，可以修改模型架构的某些参数，或者添加自定义配置项。

21. compilation_config (int or dict, 可选)

• 类型: 整数或字典，可选
• 描述: 编译配置。
• 取值:
  • 整数: 编译优化级别。
  • 字典: 完整的编译配置字典。
• 作用: 控制 vLLM 的编译优化级别和配置。 更高的编译优化级别可能会带来更好的性能，但也可能增加编译时间。 具体的编译配置细节需要参考 vLLM 的文档。

22. **kwargs (dict, 可选)

• 类型: 字典，可选
• 描述: 传递给 vllm.EngineArgs 的其他参数。
• 作用: 允许你通过 vllm.LLM 构造函数直接设置 vllm.EngineArgs 中的参数。 vllm.EngineArgs 包含了更多底层的引擎配置参数，例如 Worker 的数量、预热步数等。 详细参数请参考 vllm.EngineArgs 的文档。

---

示例配置

from vllm import LLM

# 配置一个量化模型，使用4GPU
from vllm import LLM
os.environ["CUDA_VISIBLE_DEVICES"] = "0,1,2,3"



MAX_NUM_SEQS = 16
MAX_MODEL_LEN = 8192*3//2
llm = vllm.LLM(
    "/model_path", 
    tensor_parallel_size=4,
    trust_remote_code=True,
    # enforce_eager=True,
    # dtype=torch.bfloat16,
    quantization = "awq",
    max_model_len=MAX_MODEL_LEN,
    gpu_memory_utilization=0.80, 
    max_num_seqs=MAX_NUM_SEQS,
)
tokenizer = llm.get_tokenizer()

---

EngineArgs 的参数

源码位置

https://github.com/vllm-project/vllm/blob/47532cd9f4bb751955d10989eda2078966deb0aa/vllm/engine/arg_utils.py#L91

class EngineArgs:
    """Arguments for vLLM engine."""
    model: str = 'facebook/opt-125m'
    served_model_name: Optional[Union[str, List[str]]] = None
    tokenizer: Optional[str] = None
    hf_config_path: Optional[str] = None
    task: TaskOption = "auto"
    skip_tokenizer_init: bool = False
    tokenizer_mode: str = 'auto'
    trust_remote_code: bool = False
    allowed_local_media_path: str = ""
    download_dir: Optional[str] = None
    load_format: str = 'auto'
    config_format: ConfigFormat = ConfigFormat.AUTO
    dtype: str = 'auto'
    kv_cache_dtype: str = 'auto'
    seed: Optional[int] = None
    max_model_len: Optional[int] = None
    # Note: Specifying a custom executor backend by passing a class
    # is intended for expert use only. The API may change without
    # notice.
    distributed_executor_backend: Optional[Union[str,
                                                 Type[ExecutorBase]]] = None
    # number of P/D disaggregation (or other disaggregation) workers
    pipeline_parallel_size: int = 1
    tensor_parallel_size: int = 1
    enable_expert_parallel: bool = False
    max_parallel_loading_workers: Optional[int] = None
    block_size: Optional[int] = None
    enable_prefix_caching: Optional[bool] = None
    disable_sliding_window: bool = False
    use_v2_block_manager: bool = True
    swap_space: float = 4  # GiB
    cpu_offload_gb: float = 0  # GiB
    gpu_memory_utilization: float = 0.90
    max_num_batched_tokens: Optional[int] = None
    max_num_partial_prefills: Optional[int] = 1
    max_long_partial_prefills: Optional[int] = 1
    long_prefill_token_threshold: Optional[int] = 0
    max_num_seqs: Optional[int] = None
    max_logprobs: int = 20  # Default value for OpenAI Chat Completions API
    disable_log_stats: bool = False
    revision: Optional[str] = None
    code_revision: Optional[str] = None
    rope_scaling: Optional[Dict[str, Any]] = None
    rope_theta: Optional[float] = None
    hf_overrides: Optional[HfOverrides] = None
    tokenizer_revision: Optional[str] = None
    quantization: Optional[str] = None
    enforce_eager: Optional[bool] = None
    max_seq_len_to_capture: int = 8192
    disable_custom_all_reduce: bool = False
    tokenizer_pool_size: int = 0
    # Note: Specifying a tokenizer pool by passing a class
    # is intended for expert use only. The API may change without
    # notice.
    tokenizer_pool_type: Union[str, Type["BaseTokenizerGroup"]] = "ray"
    tokenizer_pool_extra_config: Optional[Dict[str, Any]] = None
    limit_mm_per_prompt: Optional[Mapping[str, int]] = None
    mm_processor_kwargs: Optional[Dict[str, Any]] = None
    disable_mm_preprocessor_cache: bool = False
    enable_lora: bool = False
    enable_lora_bias: bool = False
    max_loras: int = 1
    max_lora_rank: int = 16
    enable_prompt_adapter: bool = False
    max_prompt_adapters: int = 1
    max_prompt_adapter_token: int = 0
    fully_sharded_loras: bool = False
    lora_extra_vocab_size: int = 256
    long_lora_scaling_factors: Optional[Tuple[float]] = None
    lora_dtype: Optional[Union[str, torch.dtype]] = 'auto'
    max_cpu_loras: Optional[int] = None
    device: str = 'auto'
    num_scheduler_steps: int = 1
    multi_step_stream_outputs: bool = True
    ray_workers_use_nsight: bool = False
    num_gpu_blocks_override: Optional[int] = None
    num_lookahead_slots: int = 0
    model_loader_extra_config: Optional[dict] = None
    ignore_patterns: Optional[Union[str, List[str]]] = None
    preemption_mode: Optional[str] = None

    scheduler_delay_factor: float = 0.0
    enable_chunked_prefill: Optional[bool] = None

    guided_decoding_backend: str = 'xgrammar'
    logits_processor_pattern: Optional[str] = None
    # Speculative decoding configuration.
    speculative_model: Optional[str] = None
    speculative_model_quantization: Optional[str] = None
    speculative_draft_tensor_parallel_size: Optional[int] = None
    num_speculative_tokens: Optional[int] = None
    speculative_disable_mqa_scorer: Optional[bool] = False
    speculative_max_model_len: Optional[int] = None
    speculative_disable_by_batch_size: Optional[int] = None
    ngram_prompt_lookup_max: Optional[int] = None
    ngram_prompt_lookup_min: Optional[int] = None
    spec_decoding_acceptance_method: str = 'rejection_sampler'
    typical_acceptance_sampler_posterior_threshold: Optional[float] = None
    typical_acceptance_sampler_posterior_alpha: Optional[float] = None
    qlora_adapter_name_or_path: Optional[str] = None
    disable_logprobs_during_spec_decoding: Optional[bool] = None

    show_hidden_metrics_for_version: Optional[str] = None
    otlp_traces_endpoint: Optional[str] = None
    collect_detailed_traces: Optional[str] = None
    disable_async_output_proc: bool = False
    scheduling_policy: Literal["fcfs", "priority"] = "fcfs"
    scheduler_cls: Union[str, Type[object]] = "vllm.core.scheduler.Scheduler"

    override_neuron_config: Optional[Dict[str, Any]] = None
    override_pooler_config: Optional[PoolerConfig] = None
    compilation_config: Optional[CompilationConfig] = None
    worker_cls: str = "auto"
    worker_extension_cls: str = ""

    kv_transfer_config: Optional[KVTransferConfig] = None

    generation_config: Optional[str] = "auto"
    override_generation_config: Optional[Dict[str, Any]] = None
    enable_sleep_mode: bool = False
    model_impl: str = "auto"

    calculate_kv_scales: Optional[bool] = None

    additional_config: Optional[Dict[str, Any]] = None
    enable_reasoning: Optional[bool] = None
    reasoning_parser: Optional[str] = None
    use_tqdm_on_load: bool = True

    def __post_init__(self):
        if not self.tokenizer:
            self.tokenizer = self.model

        # Override the default value of enable_prefix_caching if it's not set
        # by user.
        if self.enable_prefix_caching is None:
            self.enable_prefix_caching = bool(envs.VLLM_USE_V1)

        # Override max_num_seqs if it's not set by user.
        if self.max_num_seqs is None:
            self.max_num_seqs = 256 if not envs.VLLM_USE_V1 else 1024

        # support `EngineArgs(compilation_config={...})`
        # without having to manually construct a
        # CompilationConfig object
        if isinstance(self.compilation_config, (int, dict)):
            self.compilation_config = CompilationConfig.from_cli(
                str(self.compilation_config))

        # Setup plugins
        from vllm.plugins import load_general_plugins
        load_general_plugins()

    @staticmethod
    def add_cli_args(parser: FlexibleArgumentParser) -> FlexibleArgumentParser:
        """Shared CLI arguments for vLLM engine."""

        # Model arguments
        parser.add_argument(
            '--model',
            type=str,
            default=EngineArgs.model,
            help='Name or path of the huggingface model to use.')
        parser.add_argument(
            '--task',
            default=EngineArgs.task,
            choices=get_args(TaskOption),
            help='The task to use the model for. Each vLLM instance only '
            'supports one task, even if the same model can be used for '
            'multiple tasks. When the model only supports one task, ``"auto"`` '
            'can be used to select it; otherwise, you must specify explicitly '
            'which task to use.')
        parser.add_argument(
            '--tokenizer',
            type=nullable_str,
            default=EngineArgs.tokenizer,
            help='Name or path of the huggingface tokenizer to use. '
            'If unspecified, model name or path will be used.')
        parser.add_argument(
            "--hf-config-path",
            type=nullable_str,
            default=EngineArgs.hf_config_path,
            help='Name or path of the huggingface config to use. '
            'If unspecified, model name or path will be used.')
        parser.add_argument(
            '--skip-tokenizer-init',
            action='store_true',
            help='Skip initialization of tokenizer and detokenizer. '
            'Expects valid prompt_token_ids and None for prompt from '
            'the input. The generated output will contain token ids.')
        parser.add_argument(
            '--revision',
            type=nullable_str,
            default=None,
            help='The specific model version to use. It can be a branch '
            'name, a tag name, or a commit id. If unspecified, will use '
            'the default version.')
        parser.add_argument(
            '--code-revision',
            type=nullable_str,
            default=None,
            help='The specific revision to use for the model code on '
            'Hugging Face Hub. It can be a branch name, a tag name, or a '
            'commit id. If unspecified, will use the default version.')
        parser.add_argument(
            '--tokenizer-revision',
            type=nullable_str,
            default=None,
            help='Revision of the huggingface tokenizer to use. '
            'It can be a branch name, a tag name, or a commit id. '
            'If unspecified, will use the default version.')
        parser.add_argument(
            '--tokenizer-mode',
            type=str,
            default=EngineArgs.tokenizer_mode,
            choices=['auto', 'slow', 'mistral', 'custom'],
            help='The tokenizer mode.\n\n* "auto" will use the '
            'fast tokenizer if available.\n* "slow" will '
            'always use the slow tokenizer. \n* '
            '"mistral" will always use the `mistral_common` tokenizer. \n* '
            '"custom" will use --tokenizer to select the '
            'preregistered tokenizer.')
        parser.add_argument('--trust-remote-code',
                            action='store_true',
                            help='Trust remote code from huggingface.')
        parser.add_argument(
            '--allowed-local-media-path',
            type=str,
            help="Allowing API requests to read local images or videos "
            "from directories specified by the server file system. "
            "This is a security risk. "
            "Should only be enabled in trusted environments.")
        parser.add_argument('--download-dir',
                            type=nullable_str,
                            default=EngineArgs.download_dir,
                            help='Directory to download and load the weights, '
                            'default to the default cache dir of '
                            'huggingface.')
        parser.add_argument(
            '--load-format',
            type=str,
            default=EngineArgs.load_format,
            choices=[f.value for f in LoadFormat],
            help='The format of the model weights to load.\n\n'
            '* "auto" will try to load the weights in the safetensors format '
            'and fall back to the pytorch bin format if safetensors format '
            'is not available.\n'
            '* "pt" will load the weights in the pytorch bin format.\n'
            '* "safetensors" will load the weights in the safetensors format.\n'
            '* "npcache" will load the weights in pytorch format and store '
            'a numpy cache to speed up the loading.\n'
            '* "dummy" will initialize the weights with random values, '
            'which is mainly for profiling.\n'
            '* "tensorizer" will load the weights using tensorizer from '
            'CoreWeave. See the Tensorize vLLM Model script in the Examples '
            'section for more information.\n'
            '* "runai_streamer" will load the Safetensors weights using Run:ai'
            'Model Streamer \n'
            '* "bitsandbytes" will load the weights using bitsandbytes '
            'quantization.\n')
        parser.add_argument(
            '--config-format',
            default=EngineArgs.config_format,
            choices=[f.value for f in ConfigFormat],
            help='The format of the model config to load.\n\n'
            '* "auto" will try to load the config in hf format '
            'if available else it will try to load in mistral format ')
        parser.add_argument(
            '--dtype',
            type=str,
            default=EngineArgs.dtype,
            choices=[
                'auto', 'half', 'float16', 'bfloat16', 'float', 'float32'
            ],
            help='Data type for model weights and activations.\n\n'
            '* "auto" will use FP16 precision for FP32 and FP16 models, and '
            'BF16 precision for BF16 models.\n'
            '* "half" for FP16. Recommended for AWQ quantization.\n'
            '* "float16" is the same as "half".\n'
            '* "bfloat16" for a balance between precision and range.\n'
            '* "float" is shorthand for FP32 precision.\n'
            '* "float32" for FP32 precision.')
        parser.add_argument(
            '--kv-cache-dtype',
            type=str,
            choices=['auto', 'fp8', 'fp8_e5m2', 'fp8_e4m3'],
            default=EngineArgs.kv_cache_dtype,
            help='Data type for kv cache storage. If "auto", will use model '
            'data type. CUDA 11.8+ supports fp8 (=fp8_e4m3) and fp8_e5m2. '
            'ROCm (AMD GPU) supports fp8 (=fp8_e4m3)')
        parser.add_argument('--max-model-len',
                            type=int,
                            default=EngineArgs.max_model_len,
                            help='Model context length. If unspecified, will '
                            'be automatically derived from the model config.')
        parser.add_argument(
            '--guided-decoding-backend',
            type=str,
            default='xgrammar',
            help='Which engine will be used for guided decoding'
            ' (JSON schema / regex etc) by default. Currently support '
            'https://github.com/outlines-dev/outlines, '
            'https://github.com/mlc-ai/xgrammar, and '
            'https://github.com/noamgat/lm-format-enforcer.'
            ' Can be overridden per request via guided_decoding_backend'
            ' parameter.\n'
            'Backend-specific options can be supplied in a comma-separated '
            'list following a colon after the backend name. Valid backends and '
            'all available options are: [xgrammar:no-fallback, '
            'xgrammar:disable-any-whitespace, '
            'outlines:no-fallback, lm-format-enforcer:no-fallback]')
        parser.add_argument(
            '--logits-processor-pattern',
            type=nullable_str,
            default=None,
            help='Optional regex pattern specifying valid logits processor '
            'qualified names that can be passed with the `logits_processors` '
            'extra completion argument. Defaults to None, which allows no '
            'processors.')
        parser.add_argument(
            '--model-impl',
            type=str,
            default=EngineArgs.model_impl,
            choices=[f.value for f in ModelImpl],
            help='Which implementation of the model to use.\n\n'
            '* "auto" will try to use the vLLM implementation if it exists '
            'and fall back to the Transformers implementation if no vLLM '
            'implementation is available.\n'
            '* "vllm" will use the vLLM model implementation.\n'
            '* "transformers" will use the Transformers model '
            'implementation.\n')
        # Parallel arguments
        parser.add_argument(
            '--distributed-executor-backend',
            choices=['ray', 'mp', 'uni', 'external_launcher'],
            default=EngineArgs.distributed_executor_backend,
            help='Backend to use for distributed model '
            'workers, either "ray" or "mp" (multiprocessing). If the product '
            'of pipeline_parallel_size and tensor_parallel_size is less than '
            'or equal to the number of GPUs available, "mp" will be used to '
            'keep processing on a single host. Otherwise, this will default '
            'to "ray" if Ray is installed and fail otherwise. Note that tpu '
            'only supports Ray for distributed inference.')

        parser.add_argument('--pipeline-parallel-size',
                            '-pp',
                            type=int,
                            default=EngineArgs.pipeline_parallel_size,
                            help='Number of pipeline stages.')
        parser.add_argument('--tensor-parallel-size',
                            '-tp',
                            type=int,
                            default=EngineArgs.tensor_parallel_size,
                            help='Number of tensor parallel replicas.')
        parser.add_argument(
            '--enable-expert-parallel',
            action='store_true',
            help='Use expert parallelism instead of tensor parallelism '
            'for MoE layers.')
        parser.add_argument(
            '--max-parallel-loading-workers',
            type=int,
            default=EngineArgs.max_parallel_loading_workers,
            help='Load model sequentially in multiple batches, '
            'to avoid RAM OOM when using tensor '
            'parallel and large models.')
        parser.add_argument(
            '--ray-workers-use-nsight',
            action='store_true',
            help='If specified, use nsight to profile Ray workers.')
        # KV cache arguments
        parser.add_argument('--block-size',
                            type=int,
                            default=EngineArgs.block_size,
                            choices=[8, 16, 32, 64, 128],
                            help='Token block size for contiguous chunks of '
                            'tokens. This is ignored on neuron devices and '
                            'set to ``--max-model-len``. On CUDA devices, '
                            'only block sizes up to 32 are supported. '
                            'On HPU devices, block size defaults to 128.')

        parser.add_argument(
            "--enable-prefix-caching",
            action=argparse.BooleanOptionalAction,
            default=EngineArgs.enable_prefix_caching,
            help="Enables automatic prefix caching. "
            "Use ``--no-enable-prefix-caching`` to disable explicitly.",
        )
        parser.add_argument('--disable-sliding-window',
                            action='store_true',
                            help='Disables sliding window, '
                            'capping to sliding window size.')
        parser.add_argument('--use-v2-block-manager',
                            action='store_true',
                            default=True,
                            help='[DEPRECATED] block manager v1 has been '
                            'removed and SelfAttnBlockSpaceManager (i.e. '
                            'block manager v2) is now the default. '
                            'Setting this flag to True or False'
                            ' has no effect on vLLM behavior.')
        parser.add_argument(
            '--num-lookahead-slots',
            type=int,
            default=EngineArgs.num_lookahead_slots,
            help='Experimental scheduling config necessary for '
            'speculative decoding. This will be replaced by '
            'speculative config in the future; it is present '
            'to enable correctness tests until then.')

        parser.add_argument('--seed',
                            type=int,
                            default=EngineArgs.seed,
                            help='Random seed for operations.')
        parser.add_argument('--swap-space',
                            type=float,
                            default=EngineArgs.swap_space,
                            help='CPU swap space size (GiB) per GPU.')
        parser.add_argument(
            '--cpu-offload-gb',
            type=float,
            default=0,
            help='The space in GiB to offload to CPU, per GPU. '
            'Default is 0, which means no offloading. Intuitively, '
            'this argument can be seen as a virtual way to increase '
            'the GPU memory size. For example, if you have one 24 GB '
            'GPU and set this to 10, virtually you can think of it as '
            'a 34 GB GPU. Then you can load a 13B model with BF16 weight, '
            'which requires at least 26GB GPU memory. Note that this '
            'requires fast CPU-GPU interconnect, as part of the model is '
            'loaded from CPU memory to GPU memory on the fly in each '
            'model forward pass.')
        parser.add_argument(
            '--gpu-memory-utilization',
            type=float,
            default=EngineArgs.gpu_memory_utilization,
            help='The fraction of GPU memory to be used for the model '
            'executor, which can range from 0 to 1. For example, a value of '
            '0.5 would imply 50%% GPU memory utilization. If unspecified, '
            'will use the default value of 0.9. This is a per-instance '
            'limit, and only applies to the current vLLM instance.'
            'It does not matter if you have another vLLM instance running '
            'on the same GPU. For example, if you have two vLLM instances '
            'running on the same GPU, you can set the GPU memory utilization '
            'to 0.5 for each instance.')
        parser.add_argument(
            '--num-gpu-blocks-override',
            type=int,
            default=None,
            help='If specified, ignore GPU profiling result and use this number'
            ' of GPU blocks. Used for testing preemption.')
        parser.add_argument('--max-num-batched-tokens',
                            type=int,
                            default=EngineArgs.max_num_batched_tokens,
                            help='Maximum number of batched tokens per '
                            'iteration.')
        parser.add_argument(
            "--max-num-partial-prefills",
            type=int,
            default=EngineArgs.max_num_partial_prefills,
            help="For chunked prefill, the max number of concurrent \
            partial prefills."
            "Defaults to 1",
        )
        parser.add_argument(
            "--max-long-partial-prefills",
            type=int,
            default=EngineArgs.max_long_partial_prefills,
            help="For chunked prefill, the maximum number of prompts longer "
            "than --long-prefill-token-threshold that will be prefilled "
            "concurrently. Setting this less than --max-num-partial-prefills "
            "will allow shorter prompts to jump the queue in front of longer "
            "prompts in some cases, improving latency. Defaults to 1.")
        parser.add_argument(
            "--long-prefill-token-threshold",
            type=float,
            default=EngineArgs.long_prefill_token_threshold,
            help="For chunked prefill, a request is considered long if the "
            "prompt is longer than this number of tokens. Defaults to 4%% of "
            "the model's context length.",
        )
        parser.add_argument('--max-num-seqs',
                            type=int,
                            default=EngineArgs.max_num_seqs,
                            help='Maximum number of sequences per iteration.')
        parser.add_argument(
            '--max-logprobs',
            type=int,
            default=EngineArgs.max_logprobs,
            help=('Max number of log probs to return logprobs is specified in'
                  ' SamplingParams.'))
        parser.add_argument('--disable-log-stats',
                            action='store_true',
                            help='Disable logging statistics.')
        # Quantization settings.
        parser.add_argument('--quantization',
                            '-q',
                            type=nullable_str,
                            choices=[*QUANTIZATION_METHODS, None],
                            default=EngineArgs.quantization,
                            help='Method used to quantize the weights. If '
                            'None, we first check the `quantization_config` '
                            'attribute in the model config file. If that is '
                            'None, we assume the model weights are not '
                            'quantized and use `dtype` to determine the data '
                            'type of the weights.')
        parser.add_argument(
            '--rope-scaling',
            default=None,
            type=json.loads,
            help='RoPE scaling configuration in JSON format. '
            'For example, ``{"rope_type":"dynamic","factor":2.0}``')
        parser.add_argument('--rope-theta',
                            default=None,
                            type=float,
                            help='RoPE theta. Use with `rope_scaling`. In '
                            'some cases, changing the RoPE theta improves the '
                            'performance of the scaled model.')
        parser.add_argument('--hf-overrides',
                            type=json.loads,
                            default=EngineArgs.hf_overrides,
                            help='Extra arguments for the HuggingFace config. '
                            'This should be a JSON string that will be '
                            'parsed into a dictionary.')
        parser.add_argument('--enforce-eager',
                            action='store_true',
                            help='Always use eager-mode PyTorch. If False, '
                            'will use eager mode and CUDA graph in hybrid '
                            'for maximal performance and flexibility.')
        parser.add_argument('--max-seq-len-to-capture',
                            type=int,
                            default=EngineArgs.max_seq_len_to_capture,
                            help='Maximum sequence length covered by CUDA '
                            'graphs. When a sequence has context length '
                            'larger than this, we fall back to eager mode. '
                            'Additionally for encoder-decoder models, if the '
                            'sequence length of the encoder input is larger '
                            'than this, we fall back to the eager mode.')
        parser.add_argument('--disable-custom-all-reduce',
                            action='store_true',
                            default=EngineArgs.disable_custom_all_reduce,
                            help='See ParallelConfig.')
        parser.add_argument('--tokenizer-pool-size',
                            type=int,
                            default=EngineArgs.tokenizer_pool_size,
                            help='Size of tokenizer pool to use for '
                            'asynchronous tokenization. If 0, will '
                            'use synchronous tokenization.')
        parser.add_argument('--tokenizer-pool-type',
                            type=str,
                            default=EngineArgs.tokenizer_pool_type,
                            help='Type of tokenizer pool to use for '
                            'asynchronous tokenization. Ignored '
                            'if tokenizer_pool_size is 0.')
        parser.add_argument('--tokenizer-pool-extra-config',
                            type=nullable_str,
                            default=EngineArgs.tokenizer_pool_extra_config,
                            help='Extra config for tokenizer pool. '
                            'This should be a JSON string that will be '
                            'parsed into a dictionary. Ignored if '
                            'tokenizer_pool_size is 0.')

        # Multimodal related configs
        parser.add_argument(
            '--limit-mm-per-prompt',
            type=nullable_kvs,
            default=EngineArgs.limit_mm_per_prompt,
            # The default value is given in
            # MultiModalRegistry.init_mm_limits_per_prompt
            help=('For each multimodal plugin, limit how many '
                  'input instances to allow for each prompt. '
                  'Expects a comma-separated list of items, '
                  'e.g.: `image=16,video=2` allows a maximum of 16 '
                  'images and 2 videos per prompt. Defaults to 1 for '
                  'each modality.'))
        parser.add_argument(
            '--mm-processor-kwargs',
            default=None,
            type=json.loads,
            help=('Overrides for the multimodal input mapping/processing, '
                  'e.g., image processor. For example: ``{"num_crops": 4}``.'))
        parser.add_argument(
            '--disable-mm-preprocessor-cache',
            action='store_true',
            help='If true, then disables caching of the multi-modal '
            'preprocessor/mapper. (not recommended)')

        # LoRA related configs
        parser.add_argument('--enable-lora',
                            action='store_true',
                            help='If True, enable handling of LoRA adapters.')
        parser.add_argument('--enable-lora-bias',
                            action='store_true',
                            help='If True, enable bias for LoRA adapters.')
        parser.add_argument('--max-loras',
                            type=int,
                            default=EngineArgs.max_loras,
                            help='Max number of LoRAs in a single batch.')
        parser.add_argument('--max-lora-rank',
                            type=int,
                            default=EngineArgs.max_lora_rank,
                            help='Max LoRA rank.')
        parser.add_argument(
            '--lora-extra-vocab-size',
            type=int,
            default=EngineArgs.lora_extra_vocab_size,
            help=('Maximum size of extra vocabulary that can be '
                  'present in a LoRA adapter (added to the base '
                  'model vocabulary).'))
        parser.add_argument(
            '--lora-dtype',
            type=str,
            default=EngineArgs.lora_dtype,
            choices=['auto', 'float16', 'bfloat16'],
            help=('Data type for LoRA. If auto, will default to '
                  'base model dtype.'))
        parser.add_argument(
            '--long-lora-scaling-factors',
            type=nullable_str,
            default=EngineArgs.long_lora_scaling_factors,
            help=('Specify multiple scaling factors (which can '
                  'be different from base model scaling factor '
                  '- see eg. Long LoRA) to allow for multiple '
                  'LoRA adapters trained with those scaling '
                  'factors to be used at the same time. If not '
                  'specified, only adapters trained with the '
                  'base model scaling factor are allowed.'))
        parser.add_argument(
            '--max-cpu-loras',
            type=int,
            default=EngineArgs.max_cpu_loras,
            help=('Maximum number of LoRAs to store in CPU memory. '
                  'Must be >= than max_loras. '
                  'Defaults to max_loras.'))
        parser.add_argument(
            '--fully-sharded-loras',
            action='store_true',
            help=('By default, only half of the LoRA computation is '
                  'sharded with tensor parallelism. '
                  'Enabling this will use the fully sharded layers. '
                  'At high sequence length, max rank or '
                  'tensor parallel size, this is likely faster.'))
        parser.add_argument('--enable-prompt-adapter',
                            action='store_true',
                            help='If True, enable handling of PromptAdapters.')
        parser.add_argument('--max-prompt-adapters',
                            type=int,
                            default=EngineArgs.max_prompt_adapters,
                            help='Max number of PromptAdapters in a batch.')
        parser.add_argument('--max-prompt-adapter-token',
                            type=int,
                            default=EngineArgs.max_prompt_adapter_token,
                            help='Max number of PromptAdapters tokens')
        parser.add_argument("--device",
                            type=str,
                            default=EngineArgs.device,
                            choices=DEVICE_OPTIONS,
                            help='Device type for vLLM execution.')
        parser.add_argument('--num-scheduler-steps',
                            type=int,
                            default=1,
                            help=('Maximum number of forward steps per '
                                  'scheduler call.'))
        parser.add_argument(
            '--use-tqdm-on-load',
            dest='use_tqdm_on_load',
            action=argparse.BooleanOptionalAction,
            default=EngineArgs.use_tqdm_on_load,
            help='Whether to enable/disable progress bar '
            'when loading model weights.',
        )

        parser.add_argument(
            '--multi-step-stream-outputs',
            action=StoreBoolean,
            default=EngineArgs.multi_step_stream_outputs,
            nargs="?",
            const="True",
            help='If False, then multi-step will stream outputs at the end '
            'of all steps')
        parser.add_argument(
            '--scheduler-delay-factor',
            type=float,
            default=EngineArgs.scheduler_delay_factor,
            help='Apply a delay (of delay factor multiplied by previous '
            'prompt latency) before scheduling next prompt.')
        parser.add_argument(
            '--enable-chunked-prefill',
            action=StoreBoolean,
            default=EngineArgs.enable_chunked_prefill,
            nargs="?",
            const="True",
            help='If set, the prefill requests can be chunked based on the '
            'max_num_batched_tokens.')

        parser.add_argument(
            '--speculative-model',
            type=nullable_str,
            default=EngineArgs.speculative_model,
            help=
            'The name of the draft model to be used in speculative decoding.')
        # Quantization settings for speculative model.
        parser.add_argument(
            '--speculative-model-quantization',
            type=nullable_str,
            choices=[*QUANTIZATION_METHODS, None],
            default=EngineArgs.speculative_model_quantization,
            help='Method used to quantize the weights of speculative model. '
            'If None, we first check the `quantization_config` '
            'attribute in the model config file. If that is '
            'None, we assume the model weights are not '
            'quantized and use `dtype` to determine the data '
            'type of the weights.')
        parser.add_argument(
            '--num-speculative-tokens',
            type=int,
            default=EngineArgs.num_speculative_tokens,
            help='The number of speculative tokens to sample from '
            'the draft model in speculative decoding.')
        parser.add_argument(
            '--speculative-disable-mqa-scorer',
            action='store_true',
            help=
            'If set to True, the MQA scorer will be disabled in speculative '
            ' and fall back to batch expansion')
        parser.add_argument(
            '--speculative-draft-tensor-parallel-size',
            '-spec-draft-tp',
            type=int,
            default=EngineArgs.speculative_draft_tensor_parallel_size,
            help='Number of tensor parallel replicas for '
            'the draft model in speculative decoding.')

        parser.add_argument(
            '--speculative-max-model-len',
            type=int,
            default=EngineArgs.speculative_max_model_len,
            help='The maximum sequence length supported by the '
            'draft model. Sequences over this length will skip '
            'speculation.')

        parser.add_argument(
            '--speculative-disable-by-batch-size',
            type=int,
            default=EngineArgs.speculative_disable_by_batch_size,
            help='Disable speculative decoding for new incoming requests '
            'if the number of enqueue requests is larger than this value.')

        parser.add_argument(
            '--ngram-prompt-lookup-max',
            type=int,
            default=EngineArgs.ngram_prompt_lookup_max,
            help='Max size of window for ngram prompt lookup in speculative '
            'decoding.')

        parser.add_argument(
            '--ngram-prompt-lookup-min',
            type=int,
            default=EngineArgs.ngram_prompt_lookup_min,
            help='Min size of window for ngram prompt lookup in speculative '
            'decoding.')

        parser.add_argument(
            '--spec-decoding-acceptance-method',
            type=str,
            default=EngineArgs.spec_decoding_acceptance_method,
            choices=['rejection_sampler', 'typical_acceptance_sampler'],
            help='Specify the acceptance method to use during draft token '
            'verification in speculative decoding. Two types of acceptance '
            'routines are supported: '
            '1) RejectionSampler which does not allow changing the '
            'acceptance rate of draft tokens, '
            '2) TypicalAcceptanceSampler which is configurable, allowing for '
            'a higher acceptance rate at the cost of lower quality, '
            'and vice versa.')

        parser.add_argument(
            '--typical-acceptance-sampler-posterior-threshold',
            type=float,
            default=EngineArgs.typical_acceptance_sampler_posterior_threshold,
            help='Set the lower bound threshold for the posterior '
            'probability of a token to be accepted. This threshold is '
            'used by the TypicalAcceptanceSampler to make sampling decisions '
            'during speculative decoding. Defaults to 0.09')

        parser.add_argument(
            '--typical-acceptance-sampler-posterior-alpha',
            type=float,
            default=EngineArgs.typical_acceptance_sampler_posterior_alpha,
            help='A scaling factor for the entropy-based threshold for token '
            'acceptance in the TypicalAcceptanceSampler. Typically defaults '
            'to sqrt of --typical-acceptance-sampler-posterior-threshold '
            'i.e. 0.3')

        parser.add_argument(
            '--disable-logprobs-during-spec-decoding',
            action=StoreBoolean,
            default=EngineArgs.disable_logprobs_during_spec_decoding,
            nargs="?",
            const="True",
            help='If set to True, token log probabilities are not returned '
            'during speculative decoding. If set to False, log probabilities '
            'are returned according to the settings in SamplingParams. If '
            'not specified, it defaults to True. Disabling log probabilities '
            'during speculative decoding reduces latency by skipping logprob '
            'calculation in proposal sampling, target sampling, and after '
            'accepted tokens are determined.')

        parser.add_argument('--model-loader-extra-config',
                            type=nullable_str,
                            default=EngineArgs.model_loader_extra_config,
                            help='Extra config for model loader. '
                            'This will be passed to the model loader '
                            'corresponding to the chosen load_format. '
                            'This should be a JSON string that will be '
                            'parsed into a dictionary.')
        parser.add_argument(
            '--ignore-patterns',
            action="append",
            type=str,
            default=[],
            help="The pattern(s) to ignore when loading the model."
            "Default to `original/**/*` to avoid repeated loading of llama's "
            "checkpoints.")
        parser.add_argument(
            '--preemption-mode',
            type=str,
            default=None,
            help='If \'recompute\', the engine performs preemption by '
            'recomputing; If \'swap\', the engine performs preemption by '
            'block swapping.')

        parser.add_argument(
            "--served-model-name",
            nargs="+",
            type=str,
            default=None,
            help="The model name(s) used in the API. If multiple "
            "names are provided, the server will respond to any "
            "of the provided names. The model name in the model "
            "field of a response will be the first name in this "
            "list. If not specified, the model name will be the "
            "same as the ``--model`` argument. Noted that this name(s) "
            "will also be used in `model_name` tag content of "
            "prometheus metrics, if multiple names provided, metrics "
            "tag will take the first one.")
        parser.add_argument('--qlora-adapter-name-or-path',
                            type=str,
                            default=None,
                            help='Name or path of the QLoRA adapter.')

        parser.add_argument('--show-hidden-metrics-for-version',
                            type=str,
                            default=None,
                            help='Enable deprecated Prometheus metrics that '
                            'have been hidden since the specified version. '
                            'For example, if a previously deprecated metric '
                            'has been hidden since the v0.7.0 release, you '
                            'use --show-hidden-metrics-for-version=0.7 as a '
                            'temporary escape hatch while you migrate to new '
                            'metrics. The metric is likely to be removed '
                            'completely in an upcoming release.')

        parser.add_argument(
            '--otlp-traces-endpoint',
            type=str,
            default=None,
            help='Target URL to which OpenTelemetry traces will be sent.')
        parser.add_argument(
            '--collect-detailed-traces',
            type=str,
            default=None,
            help="Valid choices are " +
            ",".join(ALLOWED_DETAILED_TRACE_MODULES) +
            ". It makes sense to set this only if ``--otlp-traces-endpoint`` is"
            " set. If set, it will collect detailed traces for the specified "
            "modules. This involves use of possibly costly and or blocking "
            "operations and hence might have a performance impact.")

        parser.add_argument(
            '--disable-async-output-proc',
            action='store_true',
            default=EngineArgs.disable_async_output_proc,
            help="Disable async output processing. This may result in "
            "lower performance.")

        parser.add_argument(
            '--scheduling-policy',
            choices=['fcfs', 'priority'],
            default="fcfs",
            help='The scheduling policy to use. "fcfs" (first come first served'
            ', i.e. requests are handled in order of arrival; default) '
            'or "priority" (requests are handled based on given '
            'priority (lower value means earlier handling) and time of '
            'arrival deciding any ties).')

        parser.add_argument(
            '--scheduler-cls',
            default=EngineArgs.scheduler_cls,
            help='The scheduler class to use. "vllm.core.scheduler.Scheduler" '
            'is the default scheduler. Can be a class directly or the path to '
            'a class of form "mod.custom_class".')

        parser.add_argument(
            '--override-neuron-config',
            type=json.loads,
            default=None,
            help="Override or set neuron device configuration. "
            "e.g. ``{\"cast_logits_dtype\": \"bloat16\"}``.")
        parser.add_argument(
            '--override-pooler-config',
            type=PoolerConfig.from_json,
            default=None,
            help="Override or set the pooling method for pooling models. "
            "e.g. ``{\"pooling_type\": \"mean\", \"normalize\": false}``.")

        parser.add_argument('--compilation-config',
                            '-O',
                            type=CompilationConfig.from_cli,
                            default=None,
                            help='torch.compile configuration for the model.'
                            'When it is a number (0, 1, 2, 3), it will be '
                            'interpreted as the optimization level.\n'
                            'NOTE: level 0 is the default level without '
                            'any optimization. level 1 and 2 are for internal '
                            'testing only. level 3 is the recommended level '
                            'for production.\n'
                            'To specify the full compilation config, '
                            'use a JSON string.\n'
                            'Following the convention of traditional '
                            'compilers, using -O without space is also '
                            'supported. -O3 is equivalent to -O 3.')

        parser.add_argument('--kv-transfer-config',
                            type=KVTransferConfig.from_cli,
                            default=None,
                            help='The configurations for distributed KV cache '
                            'transfer. Should be a JSON string.')

        parser.add_argument(
            '--worker-cls',
            type=str,
            default="auto",
            help='The worker class to use for distributed execution.')
        parser.add_argument(
            '--worker-extension-cls',
            type=str,
            default="",
            help='The worker extension class on top of the worker cls, '
            'it is useful if you just want to add new functions to the worker '
            'class without changing the existing functions.')
        parser.add_argument(
            "--generation-config",
            type=nullable_str,
            default="auto",
            help="The folder path to the generation config. "
            "Defaults to 'auto', the generation config will be loaded from "
            "model path. If set to 'vllm', no generation config is loaded, "
            "vLLM defaults will be used. If set to a folder path, the "
            "generation config will be loaded from the specified folder path. "
            "If `max_new_tokens` is specified in generation config, then "
            "it sets a server-wide limit on the number of output tokens "
            "for all requests.")

        parser.add_argument(
            "--override-generation-config",
            type=json.loads,
            default=None,
            help="Overrides or sets generation config in JSON format. "
            "e.g. ``{\"temperature\": 0.5}``. If used with "
            "--generation-config=auto, the override parameters will be merged "
            "with the default config from the model. If generation-config is "
            "None, only the override parameters are used.")

        parser.add_argument("--enable-sleep-mode",
                            action="store_true",
                            default=False,
                            help="Enable sleep mode for the engine. "
                            "(only cuda platform is supported)")

        parser.add_argument(
            '--calculate-kv-scales',
            action='store_true',
            help='This enables dynamic calculation of '
            'k_scale and v_scale when kv-cache-dtype is fp8. '
            'If calculate-kv-scales is false, the scales will '
            'be loaded from the model checkpoint if available. '
            'Otherwise, the scales will default to 1.0.')

        parser.add_argument(
            "--additional-config",
            type=json.loads,
            default=None,
            help="Additional config for specified platform in JSON format. "
            "Different platforms may support different configs. Make sure the "
            "configs are valid for the platform you are using. The input format"
            " is like '{\"config_key\":\"config_value\"}'")

        parser.add_argument(
            "--enable-reasoning",
            action="store_true",
            default=False,
            help="Whether to enable reasoning_content for the model. "
            "If enabled, the model will be able to generate reasoning content."
        )

        parser.add_argument(
            "--reasoning-parser",
            type=str,
            choices=["deepseek_r1"],
            default=None,
            help=
            "Select the reasoning parser depending on the model that you're "
            "using. This is used to parse the reasoning content into OpenAI "
            "API format. Required for ``--enable-reasoning``.")

        return parser

    @classmethod
    def from_cli_args(cls, args: argparse.Namespace):
        # Get the list of attributes of this dataclass.
        attrs = [attr.name for attr in dataclasses.fields(cls)]
        # Set the attributes from the parsed arguments.
        engine_args = cls(**{attr: getattr(args, attr) for attr in attrs})
        return engine_args

    def create_model_config(self) -> ModelConfig:
        # gguf file needs a specific model loader and doesn't use hf_repo
        if check_gguf_file(self.model):
            self.quantization = self.load_format = "gguf"

        # NOTE: This is to allow model loading from S3 in CI
        if (not isinstance(self, AsyncEngineArgs) and envs.VLLM_CI_USE_S3
                and self.model in MODELS_ON_S3
                and self.load_format == LoadFormat.AUTO):  # noqa: E501
            self.model = f"{MODEL_WEIGHTS_S3_BUCKET}/{self.model}"
            self.load_format = LoadFormat.RUNAI_STREAMER

        return ModelConfig(
            model=self.model,
            hf_config_path=self.hf_config_path,
            task=self.task,
            # We know this is not None because we set it in __post_init__
            tokenizer=cast(str, self.tokenizer),
            tokenizer_mode=self.tokenizer_mode,
            trust_remote_code=self.trust_remote_code,
            allowed_local_media_path=self.allowed_local_media_path,
            dtype=self.dtype,
            seed=self.seed,
            revision=self.revision,
            code_revision=self.code_revision,
            rope_scaling=self.rope_scaling,
            rope_theta=self.rope_theta,
            hf_overrides=self.hf_overrides,
            tokenizer_revision=self.tokenizer_revision,
            max_model_len=self.max_model_len,
            quantization=self.quantization,
            enforce_eager=self.enforce_eager,
            max_seq_len_to_capture=self.max_seq_len_to_capture,
            max_logprobs=self.max_logprobs,
            disable_sliding_window=self.disable_sliding_window,
            skip_tokenizer_init=self.skip_tokenizer_init,
            served_model_name=self.served_model_name,
            limit_mm_per_prompt=self.limit_mm_per_prompt,
            use_async_output_proc=not self.disable_async_output_proc,
            config_format=self.config_format,
            mm_processor_kwargs=self.mm_processor_kwargs,
            disable_mm_preprocessor_cache=self.disable_mm_preprocessor_cache,
            override_neuron_config=self.override_neuron_config,
            override_pooler_config=self.override_pooler_config,
            logits_processor_pattern=self.logits_processor_pattern,
            generation_config=self.generation_config,
            override_generation_config=self.override_generation_config,
            enable_sleep_mode=self.enable_sleep_mode,
            model_impl=self.model_impl,
        )

    def create_load_config(self) -> LoadConfig:
        # bitsandbytes quantization needs a specific model loader
        # so we make sure the quant method and the load format are consistent
        if (self.quantization == "bitsandbytes" or
           self.qlora_adapter_name_or_path is not None) and \
           self.load_format != "bitsandbytes":
            raise ValueError(
                "BitsAndBytes quantization and QLoRA adapter only support "
                f"'bitsandbytes' load format, but got {self.load_format}")

        if (self.load_format == "bitsandbytes" or
            self.qlora_adapter_name_or_path is not None) and \
            self.quantization != "bitsandbytes":
            raise ValueError(
                "BitsAndBytes load format and QLoRA adapter only support "
                f"'bitsandbytes' quantization, but got {self.quantization}")

        return LoadConfig(
            load_format=self.load_format,
            download_dir=self.download_dir,
            model_loader_extra_config=self.model_loader_extra_config,
            ignore_patterns=self.ignore_patterns,
            use_tqdm_on_load=self.use_tqdm_on_load,
        )

    def create_engine_config(self,
                             usage_context: Optional[UsageContext] = None
                             ) -> VllmConfig:
        from vllm.platforms import current_platform
        current_platform.pre_register_and_update()

        if envs.VLLM_USE_V1:
            self._override_v1_engine_args(usage_context)

        device_config = DeviceConfig(device=self.device)
        model_config = self.create_model_config()

        if (model_config.is_multimodal_model and not envs.VLLM_USE_V1
                and self.enable_prefix_caching):
            logger.warning("--enable-prefix-caching is currently not "
                           "supported for multimodal models in v0 and "
                           "has been disabled.")
            self.enable_prefix_caching = False

        cache_config = CacheConfig(
            block_size=self.block_size,
            gpu_memory_utilization=self.gpu_memory_utilization,
            swap_space=self.swap_space,
            cache_dtype=self.kv_cache_dtype,
            is_attention_free=model_config.is_attention_free,
            num_gpu_blocks_override=self.num_gpu_blocks_override,
            sliding_window=model_config.get_sliding_window(),
            enable_prefix_caching=self.enable_prefix_caching,
            cpu_offload_gb=self.cpu_offload_gb,
            calculate_kv_scales=self.calculate_kv_scales,
        )
        parallel_config = ParallelConfig(
            pipeline_parallel_size=self.pipeline_parallel_size,
            tensor_parallel_size=self.tensor_parallel_size,
            enable_expert_parallel=self.enable_expert_parallel,
            max_parallel_loading_workers=self.max_parallel_loading_workers,
            disable_custom_all_reduce=self.disable_custom_all_reduce,
            tokenizer_pool_config=TokenizerPoolConfig.create_config(
                self.tokenizer_pool_size,
                self.tokenizer_pool_type,
                self.tokenizer_pool_extra_config,
            ),
            ray_workers_use_nsight=self.ray_workers_use_nsight,
            distributed_executor_backend=self.distributed_executor_backend,
            worker_cls=self.worker_cls,
            worker_extension_cls=self.worker_extension_cls,
        )

        max_model_len = model_config.max_model_len
        use_long_context = max_model_len > 32768
        if self.enable_chunked_prefill is None:
            # If not explicitly set, enable chunked prefill by default for
            # long context (> 32K) models. This is to avoid OOM errors in the
            # initial memory profiling phase.

            # For multimodal models and models with MLA, chunked prefill is
            # disabled by default in V0, but enabled by design in V1
            if model_config.is_multimodal_model or model_config.use_mla:
                self.enable_chunked_prefill = bool(envs.VLLM_USE_V1)

            elif use_long_context:
                is_gpu = device_config.device_type == "cuda"
                use_sliding_window = (model_config.get_sliding_window()
                                      is not None)
                use_spec_decode = self.speculative_model is not None
                from vllm.platforms import current_platform
                if (is_gpu and not use_sliding_window and not use_spec_decode
                        and not self.enable_lora
                        and not self.enable_prompt_adapter
                        and model_config.runner_type != "pooling"
                        and not current_platform.is_rocm()):
                    self.enable_chunked_prefill = True
                    logger.warning(
                        "Chunked prefill is enabled by default for models with "
                        "max_model_len > 32K. Currently, chunked prefill might "
                        "not work with some features or models. If you "
                        "encounter any issues, please disable chunked prefill "
                        "by setting --enable-chunked-prefill=False.")
            if self.enable_chunked_prefill is None:
                self.enable_chunked_prefill = False

        if not self.enable_chunked_prefill and use_long_context:
            logger.warning(
                "The model has a long context length (%s). This may cause OOM "
                "errors during the initial memory profiling phase, or result "
                "in low performance due to small KV cache space. Consider "
                "setting --max-model-len to a smaller value.", max_model_len)
        elif (self.enable_chunked_prefill
              and model_config.runner_type == "pooling"):
            msg = "Chunked prefill is not supported for pooling models"
            raise ValueError(msg)

        speculative_config = SpeculativeConfig.maybe_create_spec_config(
            target_model_config=model_config,
            target_parallel_config=parallel_config,
            target_dtype=self.dtype,
            speculative_model=self.speculative_model,
            speculative_model_quantization = \
                self.speculative_model_quantization,
            speculative_draft_tensor_parallel_size = \
                self.speculative_draft_tensor_parallel_size,
            num_speculative_tokens=self.num_speculative_tokens,
            speculative_disable_mqa_scorer=self.speculative_disable_mqa_scorer,
            speculative_disable_by_batch_size=self.
            speculative_disable_by_batch_size,
            speculative_max_model_len=self.speculative_max_model_len,
            enable_chunked_prefill=self.enable_chunked_prefill,
            disable_log_stats=self.disable_log_stats,
            ngram_prompt_lookup_max=self.ngram_prompt_lookup_max,
            ngram_prompt_lookup_min=self.ngram_prompt_lookup_min,
            draft_token_acceptance_method=\
                self.spec_decoding_acceptance_method,
            typical_acceptance_sampler_posterior_threshold=self.
            typical_acceptance_sampler_posterior_threshold,
            typical_acceptance_sampler_posterior_alpha=self.
            typical_acceptance_sampler_posterior_alpha,
            disable_logprobs=self.disable_logprobs_during_spec_decoding,
        )

        # Reminder: Please update docs/source/features/compatibility_matrix.md
        # If the feature combo become valid
        if self.num_scheduler_steps > 1:
            if speculative_config is not None:
                raise ValueError("Speculative decoding is not supported with "
                                 "multi-step (--num-scheduler-steps > 1)")
            if self.enable_chunked_prefill and self.pipeline_parallel_size > 1:
                raise ValueError("Multi-Step Chunked-Prefill is not supported "
                                 "for pipeline-parallel-size > 1")
            from vllm.platforms import current_platform
            if current_platform.is_cpu():
                logger.warning("Multi-Step (--num-scheduler-steps > 1) is "
                               "currently not supported for CPUs and has been "
                               "disabled.")
                self.num_scheduler_steps = 1

        # make sure num_lookahead_slots is set the higher value depending on
        # if we are using speculative decoding or multi-step
        num_lookahead_slots = max(self.num_lookahead_slots,
                                  self.num_scheduler_steps - 1)
        num_lookahead_slots = num_lookahead_slots \
            if speculative_config is None \
            else speculative_config.num_lookahead_slots

        scheduler_config = SchedulerConfig(
            runner_type=model_config.runner_type,
            max_num_batched_tokens=self.max_num_batched_tokens,
            max_num_seqs=self.max_num_seqs,
            max_model_len=model_config.max_model_len,
            num_lookahead_slots=num_lookahead_slots,
            delay_factor=self.scheduler_delay_factor,
            enable_chunked_prefill=self.enable_chunked_prefill,
            is_multimodal_model=model_config.is_multimodal_model,
            preemption_mode=self.preemption_mode,
            num_scheduler_steps=self.num_scheduler_steps,
            multi_step_stream_outputs=self.multi_step_stream_outputs,
            send_delta_data=(envs.VLLM_USE_RAY_SPMD_WORKER
                             and parallel_config.use_ray),
            policy=self.scheduling_policy,
            scheduler_cls=self.scheduler_cls,
            max_num_partial_prefills=self.max_num_partial_prefills,
            max_long_partial_prefills=self.max_long_partial_prefills,
            long_prefill_token_threshold=self.long_prefill_token_threshold,
        )

        lora_config = LoRAConfig(
            bias_enabled=self.enable_lora_bias,
            max_lora_rank=self.max_lora_rank,
            max_loras=self.max_loras,
            fully_sharded_loras=self.fully_sharded_loras,
            lora_extra_vocab_size=self.lora_extra_vocab_size,
            long_lora_scaling_factors=self.long_lora_scaling_factors,
            lora_dtype=self.lora_dtype,
            max_cpu_loras=self.max_cpu_loras if self.max_cpu_loras
            and self.max_cpu_loras > 0 else None) if self.enable_lora else None

        if self.qlora_adapter_name_or_path is not None and \
            self.qlora_adapter_name_or_path != "":
            if self.model_loader_extra_config is None:
                self.model_loader_extra_config = {}
            self.model_loader_extra_config[
                "qlora_adapter_name_or_path"] = self.qlora_adapter_name_or_path

        load_config = self.create_load_config()

        prompt_adapter_config = PromptAdapterConfig(
            max_prompt_adapters=self.max_prompt_adapters,
            max_prompt_adapter_token=self.max_prompt_adapter_token) \
                                        if self.enable_prompt_adapter else None

        decoding_config = DecodingConfig(
            guided_decoding_backend=self.guided_decoding_backend,
            reasoning_backend=self.reasoning_parser
            if self.enable_reasoning else None,
        )

        show_hidden_metrics = False
        if self.show_hidden_metrics_for_version is not None:
            show_hidden_metrics = version._prev_minor_version_was(
                self.show_hidden_metrics_for_version)

        detailed_trace_modules = []
        if self.collect_detailed_traces is not None:
            detailed_trace_modules = self.collect_detailed_traces.split(",")
        for m in detailed_trace_modules:
            if m not in ALLOWED_DETAILED_TRACE_MODULES:
                raise ValueError(
                    f"Invalid module {m} in collect_detailed_traces. "
                    f"Valid modules are {ALLOWED_DETAILED_TRACE_MODULES}")
        observability_config = ObservabilityConfig(
            show_hidden_metrics=show_hidden_metrics,
            otlp_traces_endpoint=self.otlp_traces_endpoint,
            collect_model_forward_time="model" in detailed_trace_modules
            or "all" in detailed_trace_modules,
            collect_model_execute_time="worker" in detailed_trace_modules
            or "all" in detailed_trace_modules,
        )

        config = VllmConfig(
            model_config=model_config,
            cache_config=cache_config,
            parallel_config=parallel_config,
            scheduler_config=scheduler_config,
            device_config=device_config,
            lora_config=lora_config,
            speculative_config=speculative_config,
            load_config=load_config,
            decoding_config=decoding_config,
            observability_config=observability_config,
            prompt_adapter_config=prompt_adapter_config,
            compilation_config=self.compilation_config,
            kv_transfer_config=self.kv_transfer_config,
            additional_config=self.additional_config,
        )

        if envs.VLLM_USE_V1:
            self._override_v1_engine_config(config)
        return config

    def _override_v1_engine_args(self, usage_context: UsageContext) -> None:
        """
        Override the EngineArgs's args based on the usage context for V1.
        """
        assert envs.VLLM_USE_V1, "V1 is not enabled"

        # V1 always uses chunked prefills.
        self.enable_chunked_prefill = True
        # V1 should use the new scheduler by default.
        # Swap it only if this arg is set to the original V0 default
        if self.scheduler_cls == EngineArgs.scheduler_cls:
            self.scheduler_cls = "vllm.v1.core.scheduler.Scheduler"

        # When no user override, set the default values based on the usage
        # context.
        # Use different default values for different hardware.

        # Try to query the device name on the current platform. If it fails,
        # it may be because the platform that imports vLLM is not the same
        # as the platform that vLLM is running on (e.g. the case of scaling
        # vLLM with Ray) and has no GPUs. In this case we use the default
        # values for non-H100/H200 GPUs.
        try:
            from vllm.platforms import current_platform
            device_name = current_platform.get_device_name().lower()
        except Exception:
            # This is only used to set default_max_num_batched_tokens
            device_name = "no-device"

        if "h100" in device_name or "h200" in device_name:
            # For H100 and H200, we use larger default values.
            default_max_num_batched_tokens = {
                UsageContext.LLM_CLASS: 16384,
                UsageContext.OPENAI_API_SERVER: 8192,
            }
        else:
            # TODO(woosuk): Tune the default values for other hardware.
            default_max_num_batched_tokens = {
                UsageContext.LLM_CLASS: 8192,
                UsageContext.OPENAI_API_SERVER: 2048,
            }

        if (self.max_num_batched_tokens is None
                and usage_context in default_max_num_batched_tokens):
            self.max_num_batched_tokens = default_max_num_batched_tokens[
                usage_context]
            logger.warning(
                "Setting max_num_batched_tokens to %d for %s usage context.",
                self.max_num_batched_tokens, usage_context.value)

    def _override_v1_engine_config(self, engine_config: VllmConfig) -> None:
        """
        Override the EngineConfig's configs based on the usage context for V1.
        """
        assert envs.VLLM_USE_V1, "V1 is not enabled"

特别说明

参数的解释使用gemini对话结果生成