ZeticMLangeHFModel

This page reflects zetic_mlange 1.8.1.

ZeticMLangeHFModel loads supported Hugging Face-hosted ZTC models through the native Android and iOS SDKs.

Import

import 'package:zetic_mlange/zetic_mlange.dart';

Constructors

`create`

static Future<ZeticMLangeHFModel> create(
  String repoId, {
  String? userAccessToken,
  String? manifestDir,
  int index = 0,
})

Parameter	Type	Default	Description
`repoId`	`String`	—	Hugging Face repository id, such as `owner/repository`.
`userAccessToken`	`String?`	`null`	Optional Hugging Face access token for private repositories.
`manifestDir`	`String?`	`null`	Optional manifest directory. Android uses this parameter; iOS accepts and ignores it.
`index`	`int`	`0`	Model index when the repository contains multiple ZTC models.

final model = await ZeticMLangeHFModel.create(
  'owner/repository',
  userAccessToken: hfToken,
  manifestDir: 'ztc',
);

Throws: MlangeException or a native SDK error if the repository, manifest, selected model index, or native runtime cannot be loaded.

Properties

`isClosed`

bool get isClosed

Returns true after close releases the native model handle.

Methods

`run`

List<Tensor> run([List<Tensor> inputs = const []])

Runs inference and returns output tensors.

Parameter	Type	Description
`inputs`	`List<Tensor>`	Input tensors matching the selected Hugging Face model's expected shapes and data types.

final outputs = model.run(inputs);

Throws: MlangeException if the input tensor count or byte length does not match the native model input buffers, or a native SDK error if inference fails.

`close`

void close()

Releases the native model handle.

After close(), calling run() or close() again throws MlangeException because the native handle has been released.

ZeticMLangeHFModel

On this page