onde_inference 1.0.0 copy "onde_inference: ^1.0.0" to clipboard
onde_inference: ^1.0.0 copied to clipboard

On-device LLM inference for Flutter & Dart. Run Qwen 2.5 models locally with Metal on iOS and macOS, CPU on Android and desktop. No cloud, no API key.

70/ 160
pub points
512
downloads

We analyzed this package 2 days ago, and awarded it 70 pub points (of a possible 160):

Passed report section
Follow Dart file conventions
30 / 30trigger folding of the section

Passed check 10/10 points: Provide a valid pubspec.yaml

Passed check 5/5 points: Provide a valid README.md

Passed check 5/5 points: Provide a valid CHANGELOG.md

Passed check 10/10 points: Use an OSI-approved license

Detected license: MIT.

Passed report section
Provide documentation
20 / 20trigger folding of the section

Passed check 10/10 points: 20% or more of the public API has dartdoc comments

117 out of 166 API elements (70.5 %) have documentation comments.

Some symbols that are missing documentation: onde_inference.ChatMessage.ChatMessage.new, onde_inference.ChatMessage.content, onde_inference.ChatMessage.role, onde_inference.EngineInfo.EngineInfo.new, onde_inference.EngineInfo.status.

Passed check 10/10 points: Package has an example

Failed report section
Platform support
0 / 20trigger folding of the section

Failed check 0/20 points: Supports 0 of 6 possible platforms (iOS, Android, Web, Windows, macOS, Linux)

Failed report section
Pass static analysis
0 / 50trigger folding of the section

Failed check 0/50 points: code has no errors, warnings, lints, or formatting issues

Found 3 issues. Showing the first 2:

ERROR: The named parameter 'toolCalls' is required, but there's no corresponding argument.

lib/src/frb_generated.dart/frb_generated.dart:929:12

    ╷
929 │     return InferenceResult(
    │            ^^^^^^^^^^^^^^^
    ╵

To reproduce make sure you are using the lints_core and run flutter analyze lib/src/frb_generated.dart/frb_generated.dart

ERROR: The named parameter 'toolCalls' is required, but there's no corresponding argument.

lib/src/frb_generated.dart/frb_generated.dart:1238:12

     ╷
1238 │     return InferenceResult(
     │            ^^^^^^^^^^^^^^^
     ╵

To reproduce make sure you are using the lints_core and run flutter analyze lib/src/frb_generated.dart/frb_generated.dart

Failed report section
Support up-to-date dependencies
20 / 40trigger folding of the section

Passed check 10/10 points: All of the package dependencies are supported in the latest version

Package Constraint Compatible Latest Notes
flutter_rust_bridge ^2.12.0 2.12.0 2.12.0
freezed_annotation ^3.1.0 3.1.0 3.1.0
Transitive dependencies
Package Constraint Compatible Latest Notes
args - 2.7.0 2.7.0
async - 2.13.1 2.13.1
build_cli_annotations - 2.1.1 2.1.1
characters - 1.4.1 1.4.1
collection - 1.19.1 1.19.1
json_annotation - 4.11.0 4.11.0
material_color_utilities - 0.13.0 0.13.0
meta - 1.17.0 1.18.2
path - 1.9.1 1.9.1
vector_math - 2.2.0 2.3.0
web - 1.1.1 1.1.1

To reproduce run dart pub outdated --no-dev-dependencies --up-to-date --no-dependency-overrides.

Passed check 10/10 points: Package supports latest stable Dart and Flutter SDKs

Failed check 0/20 points: Compatible with dependency constraint lower bounds

downgrade analysis failed failed with 3 errors:

  • MISSING_REQUIRED_ARGUMENT - lib/src/frb_generated.dart/frb_generated.dart:929:12 - The named parameter 'toolCalls' is required, but there's no corresponding argument.
  • MISSING_REQUIRED_ARGUMENT - lib/src/frb_generated.dart/frb_generated.dart:1238:12 - The named parameter 'toolCalls' is required, but there's no corresponding argument.
  • NON_ABSTRACT_CLASS_INHERITS_ABSTRACT_MEMBER - lib/src/frb_generated.dart/frb_generated.dart:1727:7 - Missing concrete implementation of 'OndeChatEngine.loadAssignedModel'.

Run flutter pub downgrade and then flutter analyze to reproduce the above problem.

You may run dart pub upgrade --tighten to update your dependency constraints, see dart.dev/go/downgrade-testing for details.

Analyzed with Pana 0.23.12, Flutter 3.41.7, Dart 3.11.5.

Check the analysis log for details.

Weekly downloads

Display as:
By versions:
7
likes
70
points
512
downloads

Documentation

API reference

Publisher

verified publisherondeinference.com

Weekly Downloads

On-device LLM inference for Flutter & Dart. Run Qwen 2.5 models locally with Metal on iOS and macOS, CPU on Android and desktop. No cloud, no API key.

Homepage
Repository (GitHub)
View/report issues

License

MIT (license)

Dependencies

flutter, flutter_rust_bridge, freezed_annotation

More

Packages that depend on onde_inference

Packages that implement onde_inference