pub async fn quantize_model( args: QuantizeArgs, _config: &Config, output_format: &str, ) -> Result<()>
Quantize model to reduce precision and size