Skip to content
pubscout

ultralytics_yolo

0.7.2, released 13 days ago by ultralytics.com

Flutter plugin for running Ultralytics YOLO computer vision models on Android and iOS.

Copyleft license
$ flutter pub add ultralytics_yolo
Source repository

What it does

Runs Ultralytics YOLO models for single-image and real-time camera inference across Android and iOS. Supports object detection, instance segmentation, semantic segmentation, depth estimation, classification, pose estimation, and oriented bounding boxes.

AI-written summary of version 0.7.2, based on the package's README and metadata. Report an inaccuracy

Use it when

  • You need to run YOLO object detection or computer vision tasks in a Flutter app
  • You want support for live camera inference using YOLOView
  • You want automatic discovery and downloading of official YOLO model assets

Skip it when

  • You require text recognition or generic machine learning tasks not suited for YOLO models
  • You want an on-device LLM package instead of computer vision

Strengths

  • Supports multiple YOLO tasks including detection, segmentation, pose, and depth estimation
  • Provides both single-image inference and real-time camera preview widgets
  • Handles automatic model discovery, download, and caching

Weaknesses

  • Requires specific native Gradle configurations and platform runtimes for optimal acceleration (like QNN)
  • Agpl-3.0 license may restrict commercial usage depending on your application structure

Setup

Kind
Plugin (native code)
Effort
Medium
Code generation
No
Direct dependencies
5

Requires platform configuration such as Android minSdkVersion 24, and optional native runtime additions if using Qualcomm NPU (QNN) models.

import 'package:ultralytics_yolo/ultralytics_yolo.dart';

final modelId = YOLO.defaultOfficialModel() ?? 'yolo26n';

YOLOView(
  modelPath: modelId,
  onResult: (results) {
    for (final r in results) {
      debugPrint('${r.className}: ${r.confidence}');
    }
  },
)

Alternatives

Packages that do the same job and could replace ultralytics_yolo.

  • google_mlkit_commons

    Use Google ML Kit for on-device machine learning tasks without custom YOLO weights.

  • tflite_flutter

    Use raw TensorFlow Lite interpreter bindings for general custom TFLite models instead of Ultralytics YOLO specifically.

Works with

Packages commonly used together with ultralytics_yolo: ones it builds on, ones built on it, and its own family.