TensorRT-LLMs/tests/unittest/api_stability/references/quant_config.yaml

methods:
  __init__:
    parameters:
      clamp_val:
        annotation: Optional[List[float]]
        default: null
      exclude_modules:
        annotation: Optional[List[str]]
        default: null
      group_size:
        annotation: int
        default: 128
      has_zero_point:
        annotation: bool
        default: false
      kv_cache_quant_algo:
        annotation: Optional[tensorrt_llm.quantization.mode.QuantAlgo]
        default: null
      mamba_ssm_cache_dtype:
        annotation: Optional[str]
        default: null
      pre_quant_scale:
        annotation: bool
        default: false
      quant_algo:
        annotation: Optional[tensorrt_llm.quantization.mode.QuantAlgo]
        default: null
      smoothquant_val:
        annotation: float
        default: 0.5
      use_meta_recipe:
        annotation: bool
        default: false
    return_annotation: None
  from_dict:
    parameters:
      config:
        annotation: dict
        default: inspect._empty
    return_annotation: tensorrt_llm.models.modeling_utils.QuantConfig
  to_dict:
    parameters: {}
    return_annotation: dict
  is_module_excluded_from_quantization:
    parameters:
      name:
        annotation: str
        default: inspect._empty
    return_annotation: bool
properties: {}