Huggingface.js documentation

Interface: MlxQuantizationConfig

Hugging Face's logo
Join the Hugging Face community

and get access to the augmented documentation experience

to get started

Interface: MlxQuantizationConfig

Hierarchy

Indexable

▪ [key: string]: unknown

Properties

bits

• Optional bits: number

Inherited from

QuantizationConfig.bits

Defined in

packages/hub/src/lib/parse-safetensors-metadata.ts:773


config _ groups

• Optional config_groups: Record<string, { format?: string ; targets?: string[] ; weights?: { num_bits?: number } }>

Inherited from

QuantizationConfig.config_groups

Defined in

packages/hub/src/lib/parse-safetensors-metadata.ts:778


expert _ dtype

• Optional expert_dtype: string

Routed expert precision when it differs from the main quantizer (e.g. FP4 experts with FP8 attention).

Inherited from

QuantizationConfig.expert_dtype

Defined in

packages/hub/src/lib/parse-safetensors-metadata.ts:763


format

• Optional format: string

Inherited from

QuantizationConfig.format

Defined in

packages/hub/src/lib/parse-safetensors-metadata.ts:777


group _ size

• Optional group_size: number

Inherited from

QuantizationConfig.group_size

Defined in

packages/hub/src/lib/parse-safetensors-metadata.ts:771


ignore

• Optional ignore: string[]

compressed-tensors names its exclusion list ignore rather than modules_to_not_convert, using the same re:-prefixed target syntax as config_groups[].targets.

Inherited from

QuantizationConfig.ignore

Defined in

packages/hub/src/lib/parse-safetensors-metadata.ts:783


load _ in _ 4bit

• Optional load_in_4bit: boolean

Inherited from

QuantizationConfig.load_in_4bit

Defined in

packages/hub/src/lib/parse-safetensors-metadata.ts:774


load _ in _ 8bit

• Optional load_in_8bit: boolean

Inherited from

QuantizationConfig.load_in_8bit

Defined in

packages/hub/src/lib/parse-safetensors-metadata.ts:775


mode

• Optional mode: string

MLX quantization mode (e.g. affine); MLX configs do not declare quant_method.

Inherited from

QuantizationConfig.mode

Defined in

packages/hub/src/lib/parse-safetensors-metadata.ts:770


modules _ to _ not _ convert

• Optional modules_to_not_convert: string[]

Inherited from

QuantizationConfig.modules_to_not_convert

Defined in

packages/hub/src/lib/parse-safetensors-metadata.ts:772


quant _ method

• Optional quant_method: string

Inherited from

QuantizationConfig.quant_method

Defined in

packages/hub/src/lib/parse-safetensors-metadata.ts:761


store _ dtype

• Optional store_dtype: string

Same role as expert_dtype under another name: MiMo-V2.6 is quant_method: "fp8" for its dense layers but stores the routed experts as store_dtype: "mxfp4", packed two per U8.

Inherited from

QuantizationConfig.store_dtype

Defined in

packages/hub/src/lib/parse-safetensors-metadata.ts:768

Update on GitHub