Skip to content
pubscout

Compare

The strongest value in each row is marked.

Attributeflutter_litert

LiteRT (formerly TensorFlow Lite) Flutter plugin. Drop-in on-device ML inference with bundled native libraries for supported native platforms and web runtimes.

pub.devRemove
flutter_onnxruntime

A native wrapper plugin for running ONNX Runtime models across multiple platforms.

pub.devRemove
ScoresHealth86 (best)79
Maintenance97 (best)77
Quality9396 (best)
Popularity7179 (best)
Community63 (best)52
Compatibility100 (best)85
UsageDownloads (30 days)20k46k (best)
Likes3453 (best)
Pub points160 / 160 (best)150 / 160
GitHub stars2167 (best)
MaintenanceLatest release4.2.0, 5 days ago (best)1.9.0, 7 weeks ago
Major versions in 12 months20 (best)
Contributors5 (best)3
Publisherhugo.mlmasic.ai
FitPlatformsAndroid, iOS, Web, Windows, macOS, LinuxAndroid, iOS, Web, Windows, macOS, Linux
WASM-readyYesNo
KindPlugin (native code)Plugin (native code)
Code generationNoNo
Direct dependencies43 (best)
LicenseAPACHE-2.0MIT
Setup effort—Medium
JudgementStrengthsNo summary yet
  • Supports multiple platforms (Android, iOS, web, Windows, macOS, Linux).
  • Fetches binaries directly from official repositories so libraries are up-to-date.
  • Handles memory management in native code to reduce leaks.
WeaknessesNo summary yet
  • Retrieving model metadata and input/output info is unavailable on Swift and Javascript APIs.
  • Requires specific manual configuration for Android ProGuard and OS minimum versions on iOS/macOS.
Warnings
  • Native setup
  • Native setup
flutter_litert vs flutter_onnxruntime: which Flutter package should you use? | pubscout