diff --git a/packages/firebase_ml_vision/CHANGELOG.md b/packages/firebase_ml_vision/CHANGELOG.md
index 058b0d053560..f25d393d393f 100644
--- a/packages/firebase_ml_vision/CHANGELOG.md
+++ b/packages/firebase_ml_vision/CHANGELOG.md
@@ -1,3 +1,7 @@
+## 0.9.1
+
+* Add support for cloud text recognizer.
+
 ## 0.9.0+3
 
 * Automatically use version from pubspec.yaml when reporting usage to Firebase.
diff --git a/packages/firebase_ml_vision/android/src/main/java/io/flutter/plugins/firebasemlvision/TextRecognizer.java b/packages/firebase_ml_vision/android/src/main/java/io/flutter/plugins/firebasemlvision/TextRecognizer.java
index 55aa3c8bc5d7..9a0f82343ff4 100644
--- a/packages/firebase_ml_vision/android/src/main/java/io/flutter/plugins/firebasemlvision/TextRecognizer.java
+++ b/packages/firebase_ml_vision/android/src/main/java/io/flutter/plugins/firebasemlvision/TextRecognizer.java
@@ -21,7 +21,15 @@ public class TextRecognizer implements Detector {
   private final FirebaseVisionTextRecognizer recognizer;
 
   TextRecognizer(FirebaseVision vision, Map<String, Object> options) {
-    recognizer = vision.getOnDeviceTextRecognizer();
+    final String modelType = (String) options.get("modelType");
+    if (modelType.equals("onDevice")) {
+      recognizer = vision.getOnDeviceTextRecognizer();
+    } else if (modelType.equals("cloud")) {
+      recognizer = vision.getCloudTextRecognizer();
+    } else {
+      final String message = String.format("No model for type: %s", modelType);
+      throw new IllegalArgumentException(message);
+    }
   }
 
   @Override
diff --git a/packages/firebase_ml_vision/example/lib/camera_preview_scanner.dart b/packages/firebase_ml_vision/example/lib/camera_preview_scanner.dart
index c8c5eedf7a64..f40e20d6591c 100644
--- a/packages/firebase_ml_vision/example/lib/camera_preview_scanner.dart
+++ b/packages/firebase_ml_vision/example/lib/camera_preview_scanner.dart
@@ -29,6 +29,8 @@ class _CameraPreviewScannerState extends State<CameraPreviewScanner> {
   final ImageLabeler _cloudImageLabeler =
       FirebaseVision.instance.cloudImageLabeler();
   final TextRecognizer _recognizer = FirebaseVision.instance.textRecognizer();
+  final TextRecognizer _cloudRecognizer =
+      FirebaseVision.instance.cloudTextRecognizer();
 
   @override
   void initState() {
@@ -72,6 +74,8 @@ class _CameraPreviewScannerState extends State<CameraPreviewScanner> {
     switch (_currentDetector) {
       case Detector.text:
         return _recognizer.processImage;
+      case Detector.cloudText:
+        return _cloudRecognizer.processImage;
       case Detector.barcode:
         return _barcodeDetector.detectInImage;
       case Detector.label:
@@ -119,7 +123,8 @@ class _CameraPreviewScannerState extends State<CameraPreviewScanner> {
         painter = LabelDetectorPainter(imageSize, _scanResults);
         break;
       default:
-        assert(_currentDetector == Detector.text);
+        assert(_currentDetector == Detector.text ||
+            _currentDetector == Detector.cloudText);
         if (_scanResults is! VisionText) return noResultsText;
         painter = TextDetectorPainter(imageSize, _scanResults);
     }
@@ -200,6 +205,10 @@ class _CameraPreviewScannerState extends State<CameraPreviewScanner> {
                 child: Text('Detect Text'),
                 value: Detector.text,
               ),
+              const PopupMenuItem<Detector>(
+                child: Text('Detect Cloud Text'),
+                value: Detector.cloudText,
+              ),
             ],
           ),
         ],
@@ -222,6 +231,7 @@ class _CameraPreviewScannerState extends State<CameraPreviewScanner> {
       _imageLabeler.close();
       _cloudImageLabeler.close();
       _recognizer.close();
+      _cloudRecognizer.close();
     });
 
     _currentDetector = null;
diff --git a/packages/firebase_ml_vision/example/lib/detector_painters.dart b/packages/firebase_ml_vision/example/lib/detector_painters.dart
index 684503b1086c..d8d7316f2509 100644
--- a/packages/firebase_ml_vision/example/lib/detector_painters.dart
+++ b/packages/firebase_ml_vision/example/lib/detector_painters.dart
@@ -7,7 +7,7 @@ import 'dart:ui' as ui;
 import 'package:firebase_ml_vision/firebase_ml_vision.dart';
 import 'package:flutter/material.dart';
 
-enum Detector { barcode, face, label, cloudLabel, text }
+enum Detector { barcode, face, label, cloudLabel, text, cloudText }
 
 class BarcodeDetectorPainter extends CustomPainter {
   BarcodeDetectorPainter(this.absoluteImageSize, this.barcodeLocations);
diff --git a/packages/firebase_ml_vision/example/lib/picture_scanner.dart b/packages/firebase_ml_vision/example/lib/picture_scanner.dart
index 88b1c5a1f0bb..182a55a46020 100644
--- a/packages/firebase_ml_vision/example/lib/picture_scanner.dart
+++ b/packages/firebase_ml_vision/example/lib/picture_scanner.dart
@@ -29,6 +29,8 @@ class _PictureScannerState extends State<PictureScanner> {
   final ImageLabeler _cloudImageLabeler =
       FirebaseVision.instance.cloudImageLabeler();
   final TextRecognizer _recognizer = FirebaseVision.instance.textRecognizer();
+  final TextRecognizer _cloudRecognizer =
+      FirebaseVision.instance.cloudTextRecognizer();
 
   Future<void> _getAndScanImage() async {
     setState(() {
@@ -93,6 +95,9 @@ class _PictureScannerState extends State<PictureScanner> {
       case Detector.text:
         results = await _recognizer.processImage(visionImage);
         break;
+      case Detector.cloudText:
+        results = await _cloudRecognizer.processImage(visionImage);
+        break;
       default:
         return;
     }
@@ -121,6 +126,9 @@ class _PictureScannerState extends State<PictureScanner> {
       case Detector.text:
         painter = TextDetectorPainter(_imageSize, results);
         break;
+      case Detector.cloudText:
+        painter = TextDetectorPainter(_imageSize, results);
+        break;
       default:
         break;
     }
@@ -185,6 +193,10 @@ class _PictureScannerState extends State<PictureScanner> {
                 child: Text('Detect Text'),
                 value: Detector.text,
               ),
+              const PopupMenuItem<Detector>(
+                child: Text('Detect Cloud Text'),
+                value: Detector.cloudText,
+              ),
             ],
           ),
         ],
@@ -207,6 +219,7 @@ class _PictureScannerState extends State<PictureScanner> {
     _imageLabeler.close();
     _cloudImageLabeler.close();
     _recognizer.close();
+    _cloudRecognizer.close();
     super.dispose();
   }
 }
diff --git a/packages/firebase_ml_vision/ios/Classes/TextRecognizer.m b/packages/firebase_ml_vision/ios/Classes/TextRecognizer.m
index fa9bdd3158e2..cc8cc9306a12 100644
--- a/packages/firebase_ml_vision/ios/Classes/TextRecognizer.m
+++ b/packages/firebase_ml_vision/ios/Classes/TextRecognizer.m
@@ -8,7 +8,17 @@ @implementation TextRecognizer
 - (instancetype)initWithVision:(FIRVision *)vision options:(NSDictionary *)options {
   self = [super init];
   if (self) {
-    _recognizer = [vision onDeviceTextRecognizer];
+    if ([@"onDevice" isEqualToString:options[@"modelType"]]) {
+      _recognizer = [vision onDeviceTextRecognizer];
+    } else if ([@"cloud" isEqualToString:options[@"modelType"]]) {
+      _recognizer = [vision cloudTextRecognizer];
+    } else {
+      NSString *reason =
+          [NSString stringWithFormat:@"Invalid model type: %@", options[@"modelType"]];
+      @throw [[NSException alloc] initWithName:NSInvalidArgumentException
+                                        reason:reason
+                                      userInfo:nil];
+    }
   }
   return self;
 }
diff --git a/packages/firebase_ml_vision/lib/src/firebase_vision.dart b/packages/firebase_ml_vision/lib/src/firebase_vision.dart
index 6f6b0446e3ea..d0f593cff8f8 100644
--- a/packages/firebase_ml_vision/lib/src/firebase_vision.dart
+++ b/packages/firebase_ml_vision/lib/src/firebase_vision.dart
@@ -67,7 +67,12 @@ class FirebaseVision {
   }
 
   /// Creates an instance of [TextRecognizer].
-  TextRecognizer textRecognizer() => TextRecognizer._(nextHandle++);
+  TextRecognizer textRecognizer() {
+    return TextRecognizer._(
+      modelType: ModelType.onDevice,
+      handle: nextHandle++,
+    );
+  }
 
   /// Creates a cloud instance of [ImageLabeler].
   ImageLabeler cloudImageLabeler([CloudImageLabelerOptions options]) {
@@ -77,6 +82,14 @@ class FirebaseVision {
       handle: nextHandle++,
     );
   }
+
+  /// Creates a cloud instance of [TextRecognizer].
+  TextRecognizer cloudTextRecognizer() {
+    return TextRecognizer._(
+      modelType: ModelType.cloud,
+      handle: nextHandle++,
+    );
+  }
 }
 
 /// Represents an image object used for both on-device and cloud API detectors.
diff --git a/packages/firebase_ml_vision/lib/src/text_recognizer.dart b/packages/firebase_ml_vision/lib/src/text_recognizer.dart
index 888f5fd5e03b..98936f66a906 100644
--- a/packages/firebase_ml_vision/lib/src/text_recognizer.dart
+++ b/packages/firebase_ml_vision/lib/src/text_recognizer.dart
@@ -19,7 +19,13 @@ part of firebase_ml_vision;
 ///     await textRecognizer.processImage(image);
 /// ```
 class TextRecognizer {
-  TextRecognizer._(this._handle);
+  TextRecognizer._({
+    @required this.modelType,
+    @required int handle,
+  })  : _handle = handle,
+        assert(modelType != null);
+
+  final ModelType modelType;
 
   final int _handle;
   bool _hasBeenOpened = false;
@@ -35,7 +41,9 @@ class TextRecognizer {
       'TextRecognizer#processImage',
       <String, dynamic>{
         'handle': _handle,
-        'options': <String, dynamic>{},
+        'options': <String, dynamic>{
+          'modelType': _enumToString(modelType),
+        },
       }..addAll(visionImage._serialize()),
     );
 
diff --git a/packages/firebase_ml_vision/pubspec.yaml b/packages/firebase_ml_vision/pubspec.yaml
index 765b72b16a09..c2138b1866d0 100644
--- a/packages/firebase_ml_vision/pubspec.yaml
+++ b/packages/firebase_ml_vision/pubspec.yaml
@@ -2,7 +2,7 @@ name: firebase_ml_vision
 description: Flutter plugin for Firebase machine learning vision services.
 author: Flutter Team <flutter-dev@googlegroups.com>
 homepage: https://github.com/flutter/plugins/tree/master/packages/firebase_ml_vision
-version: 0.9.0+3
+version: 0.9.1
 
 dependencies:
   flutter:
diff --git a/packages/firebase_ml_vision/test/firebase_ml_vision_test.dart b/packages/firebase_ml_vision/test/firebase_ml_vision_test.dart
index 9420ddad85a0..85e069774b9a 100644
--- a/packages/firebase_ml_vision/test/firebase_ml_vision_test.dart
+++ b/packages/firebase_ml_vision/test/firebase_ml_vision_test.dart
@@ -83,7 +83,9 @@ void main() {
                   },
                 ],
               },
-              'options': <String, dynamic>{},
+              'options': <String, dynamic>{
+                'modelType': 'onDevice',
+              },
             },
           ),
         ]);
@@ -778,7 +780,268 @@ void main() {
           expect(block.confidence, 0.5);
 
           block = text.blocks[1];
+          // TODO(jackson): Use const Rect when available in minimum Flutter SDK
+          // ignore: prefer_const_constructors
+          expect(block.boundingBox, Rect.fromLTWH(14.0, 13.0, 16.0, 15.0));
+          expect(block.text, 'hello');
+          expect(block.cornerPoints, const <Offset>[
+            Offset(18.0, 17.0),
+            Offset(20.0, 19.0),
+          ]);
+          expect(block.confidence, 0.6);
+        });
+      });
+
+      group('$TextLine', () {
+        test('processImage', () async {
+          final VisionText text = await recognizer.processImage(image);
+
+          TextLine line = text.blocks[0].lines[0];
+          // TODO(jackson): Use const Rect when available in minimum Flutter SDK
+          // ignore: prefer_const_constructors
+          expect(line.boundingBox, Rect.fromLTWH(5, 6, 7, 8));
+          expect(line.text, 'friend');
+          expect(line.cornerPoints, const <Offset>[
+            Offset(9.0, 10.0),
+            Offset(11.0, 12.0),
+          ]);
+          expect(line.recognizedLanguages, hasLength(2));
+          expect(line.recognizedLanguages[0].languageCode, 'ef');
+          expect(line.recognizedLanguages[1].languageCode, 'gh');
+          expect(line.confidence, 0.3);
+
+          line = text.blocks[0].lines[1];
+          // TODO(jackson): Use const Rect when available in minimum Flutter SDK
+          // ignore: prefer_const_constructors
+          expect(line.boundingBox, Rect.fromLTWH(8.0, 7.0, 4.0, 5.0));
+          expect(line.text, 'how');
+          expect(line.cornerPoints, const <Offset>[
+            Offset(10.0, 9.0),
+            Offset(12.0, 11.0),
+          ]);
+          expect(line.confidence, 0.4);
+        });
+      });
+
+      group('$TextElement', () {
+        test('processImage', () async {
+          final VisionText text = await recognizer.processImage(image);
+
+          TextElement element = text.blocks[0].lines[0].elements[0];
           // ignore: prefer_const_constructors
+          expect(element.boundingBox, Rect.fromLTWH(1.0, 2.0, 3.0, 4.0));
+          expect(element.text, 'hello');
+          expect(element.cornerPoints, const <Offset>[
+            Offset(5.0, 6.0),
+            Offset(7.0, 8.0),
+          ]);
+          expect(element.recognizedLanguages, hasLength(2));
+          expect(element.recognizedLanguages[0].languageCode, 'ab');
+          expect(element.recognizedLanguages[1].languageCode, 'cd');
+          expect(element.confidence, 0.1);
+
+          element = text.blocks[0].lines[0].elements[1];
+          // TODO(jackson): Use const Rect when available in minimum Flutter SDK
+          // ignore: prefer_const_constructors
+          expect(element.boundingBox, Rect.fromLTWH(4.0, 3.0, 2.0, 1.0));
+          expect(element.text, 'my');
+          expect(element.cornerPoints, const <Offset>[
+            Offset(6.0, 5.0),
+            Offset(8.0, 7.0),
+          ]);
+          expect(element.confidence, 0.2);
+        });
+      });
+
+      test('processImage', () async {
+        final VisionText text = await recognizer.processImage(image);
+
+        expect(text.text, 'testext');
+        expect(log, <Matcher>[
+          isMethodCall(
+            'TextRecognizer#processImage',
+            arguments: <String, dynamic>{
+              'handle': 0,
+              'type': 'file',
+              'path': 'empty',
+              'bytes': null,
+              'metadata': null,
+              'options': <String, dynamic>{
+                'modelType': 'onDevice',
+              },
+            },
+          ),
+        ]);
+      });
+
+      test('processImage no bounding box', () async {
+        returnValue = <dynamic, dynamic>{
+          'blocks': <dynamic>[
+            <dynamic, dynamic>{
+              'text': '',
+              'points': <dynamic>[],
+              'recognizedLanguages': <dynamic>[],
+              'lines': <dynamic>[],
+            },
+          ],
+        };
+
+        final VisionText text = await recognizer.processImage(image);
+
+        final TextBlock block = text.blocks[0];
+        expect(block.boundingBox, null);
+      });
+    });
+
+    group('Cloud $TextRecognizer', () {
+      TextRecognizer recognizer;
+      final FirebaseVisionImage image = FirebaseVisionImage.fromFilePath(
+        'empty',
+      );
+
+      setUp(() {
+        recognizer = FirebaseVision.instance.cloudTextRecognizer();
+        final List<dynamic> elements = <dynamic>[
+          <dynamic, dynamic>{
+            'text': 'hello',
+            'left': 1.0,
+            'top': 2.0,
+            'width': 3.0,
+            'height': 4.0,
+            'points': <dynamic>[
+              <dynamic>[5.0, 6.0],
+              <dynamic>[7.0, 8.0],
+            ],
+            'recognizedLanguages': <dynamic>[
+              <dynamic, dynamic>{
+                'languageCode': 'ab',
+              },
+              <dynamic, dynamic>{
+                'languageCode': 'cd',
+              }
+            ],
+            'confidence': 0.1,
+          },
+          <dynamic, dynamic>{
+            'text': 'my',
+            'left': 4.0,
+            'top': 3.0,
+            'width': 2.0,
+            'height': 1.0,
+            'points': <dynamic>[
+              <dynamic>[6.0, 5.0],
+              <dynamic>[8.0, 7.0],
+            ],
+            'recognizedLanguages': <dynamic>[],
+            'confidence': 0.2,
+          },
+        ];
+
+        final List<dynamic> lines = <dynamic>[
+          <dynamic, dynamic>{
+            'text': 'friend',
+            'left': 5.0,
+            'top': 6.0,
+            'width': 7.0,
+            'height': 8.0,
+            'points': <dynamic>[
+              <dynamic>[9.0, 10.0],
+              <dynamic>[11.0, 12.0],
+            ],
+            'recognizedLanguages': <dynamic>[
+              <dynamic, dynamic>{
+                'languageCode': 'ef',
+              },
+              <dynamic, dynamic>{
+                'languageCode': 'gh',
+              }
+            ],
+            'elements': elements,
+            'confidence': 0.3,
+          },
+          <dynamic, dynamic>{
+            'text': 'how',
+            'left': 8.0,
+            'top': 7.0,
+            'width': 4.0,
+            'height': 5.0,
+            'points': <dynamic>[
+              <dynamic>[10.0, 9.0],
+              <dynamic>[12.0, 11.0],
+            ],
+            'recognizedLanguages': <dynamic>[],
+            'elements': <dynamic>[],
+            'confidence': 0.4,
+          },
+        ];
+
+        final List<dynamic> blocks = <dynamic>[
+          <dynamic, dynamic>{
+            'text': 'friend',
+            'left': 13.0,
+            'top': 14.0,
+            'width': 15.0,
+            'height': 16.0,
+            'points': <dynamic>[
+              <dynamic>[17.0, 18.0],
+              <dynamic>[19.0, 20.0],
+            ],
+            'recognizedLanguages': <dynamic>[
+              <dynamic, dynamic>{
+                'languageCode': 'ij',
+              },
+              <dynamic, dynamic>{
+                'languageCode': 'kl',
+              }
+            ],
+            'lines': lines,
+            'confidence': 0.5,
+          },
+          <dynamic, dynamic>{
+            'text': 'hello',
+            'left': 14.0,
+            'top': 13.0,
+            'width': 16.0,
+            'height': 15.0,
+            'points': <dynamic>[
+              <dynamic>[18.0, 17.0],
+              <dynamic>[20.0, 19.0],
+            ],
+            'recognizedLanguages': <dynamic>[],
+            'lines': <dynamic>[],
+            'confidence': 0.6,
+          },
+        ];
+
+        final dynamic visionText = <dynamic, dynamic>{
+          'text': 'testext',
+          'blocks': blocks,
+        };
+
+        returnValue = visionText;
+      });
+
+      group('$TextBlock', () {
+        test('processImage', () async {
+          final VisionText text = await recognizer.processImage(image);
+
+          expect(text.blocks, hasLength(2));
+
+          TextBlock block = text.blocks[0];
+          // TODO(jackson): Use const Rect when available in minimum Flutter SDK
+          // ignore: prefer_const_constructors
+          expect(block.boundingBox, Rect.fromLTWH(13.0, 14.0, 15.0, 16.0));
+          expect(block.text, 'friend');
+          expect(block.cornerPoints, const <Offset>[
+            Offset(17.0, 18.0),
+            Offset(19.0, 20.0),
+          ]);
+          expect(block.recognizedLanguages, hasLength(2));
+          expect(block.recognizedLanguages[0].languageCode, 'ij');
+          expect(block.recognizedLanguages[1].languageCode, 'kl');
+          expect(block.confidence, 0.5);
+
+          block = text.blocks[1];
           // TODO(jackson): Use const Rect when available in minimum Flutter SDK
           // ignore: prefer_const_constructors
           expect(block.boundingBox, Rect.fromLTWH(14.0, 13.0, 16.0, 15.0));
@@ -827,6 +1090,7 @@ void main() {
           final VisionText text = await recognizer.processImage(image);
 
           TextElement element = text.blocks[0].lines[0].elements[0];
+          // TODO(jackson): Use const Rect when available in minimum Flutter SDK
           // ignore: prefer_const_constructors
           expect(element.boundingBox, Rect.fromLTWH(1.0, 2.0, 3.0, 4.0));
           expect(element.text, 'hello');
@@ -865,7 +1129,9 @@ void main() {
               'path': 'empty',
               'bytes': null,
               'metadata': null,
-              'options': <String, dynamic>{},
+              'options': <String, dynamic>{
+                'modelType': 'cloud',
+              },
             },
           ),
         ]);