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

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