Skip to main content

model_quantize

Function model_quantize 

Source
pub fn model_quantize(
    fname_inp: &str,
    fname_out: &str,
    params: Option<&llama_model_quantize_params>,
) -> u32
Expand description

Quantize a model file.

§Parameters

  • fname_inp: Path to the input model file.
  • fname_out: Path to the output quantized model file.
  • params: Quantization parameters. Use None for defaults.

§Returns

Returns 0 on success, non-zero on failure.

§Panics

Panics if the paths contain null bytes.