--- description: Capacitor ML Kit 8.2.0 adds 14 new plugins for on-device machine learning, including six GenAI plugins powered by Gemini Nano on Android. title: Capacitor ML Kit 8.2.0: 14 New Plugins - Capawesome image: https://capawesome.io/docs/assets/images/social/blog/capacitor-mlkit-8-2-0-release.png --- [Skip to content ](#capacitor-ml-kit-820-14-new-plugins) [🖥️ Introducing the **Capacitor Electron Platform** — build desktop apps for macOS, Windows, and Linux. Free & open source. ](/blog/announcing-the-capacitor-electron-platform/) * [ SDKs ](/docs/sdks/) * [ Formbricks ](/docs/sdks/capacitor/formbricks/) * [ Geocoder ](/docs/sdks/capacitor/geocoder/) * [ Google Sign-In ](/docs/sdks/capacitor/google-sign-in/) * [ Grafana Faro ](/docs/sdks/capacitor/grafana-faro/) * [ Gyroscope ](/docs/sdks/capacitor/gyroscope/) * [ Haptics ](/docs/sdks/capacitor/haptics/) * [ Home Indicator ](/docs/sdks/capacitor/home-indicator/) * [ In-App Browser ](/docs/sdks/capacitor/in-app-browser/) * [ Install Referrer ](/docs/sdks/capacitor/install-referrer/) * [ Intercom ](/docs/sdks/capacitor/intercom/) * [ Intune ](/docs/sdks/capacitor/intune/) * [ Keep Awake ](/docs/sdks/capacitor/keep-awake/) * [ libSQL ](/docs/sdks/capacitor/libsql/) * [ Light Sensor ](/docs/sdks/capacitor/light-sensor/) * [ Live Update ](/docs/sdks/capacitor/live-update/) * [ Localization ](/docs/sdks/capacitor/localization/) * [ Mail Composer ](/docs/sdks/capacitor/mail-composer/) * [ Managed Configurations ](/docs/sdks/capacitor/managed-configurations/) * [ Maps Launcher ](/docs/sdks/capacitor/maps-launcher/) * [ Media Session ](/docs/sdks/capacitor/media-session/) * [ ML Kit ](/docs/sdks/capacitor/mlkit/) * [ Navigation Bar ](/docs/sdks/capacitor/navigation-bar/) * [ Network ](/docs/sdks/capacitor/network/) * [ NFC ](/docs/sdks/capacitor/nfc/) * [ Node.js ](/docs/sdks/capacitor/nodejs/) * [ OAuth ](/docs/sdks/capacitor/oauth/) * [ Passkeys ](/docs/sdks/capacitor/passkeys/) * [ Password Autofill ](/docs/sdks/capacitor/password-autofill/) * [ PDF Generator ](/docs/sdks/capacitor/pdf-generator/) * [ PDF Viewer ](/docs/sdks/capacitor/pdf-viewer/) * [ Pedometer ](/docs/sdks/capacitor/pedometer/) * [ Permissions ](/docs/sdks/capacitor/permissions/) * [ Phone Dialer ](/docs/sdks/capacitor/phone-dialer/) * [ Photo Editor ](/docs/sdks/capacitor/photo-editor/) * [ Photo Manipulator ](/docs/sdks/capacitor/photo-manipulator/) * [ PixLive ](/docs/sdks/capacitor/pixlive/) * [ PostHog ](/docs/sdks/capacitor/posthog/) * [ Printer ](/docs/sdks/capacitor/printer/) * [ Privacy Screen ](/docs/sdks/capacitor/privacy-screen/) * [ Proximity Sensor ](/docs/sdks/capacitor/proximity-sensor/) * [ Purchases ](/docs/sdks/capacitor/purchases/) * [ RealtimeKit ](/docs/sdks/capacitor/realtimekit/) * [ Root Detection ](/docs/sdks/capacitor/root-detection/) * [ Screen Brightness ](/docs/sdks/capacitor/screen-brightness/) * [ Screen Orientation ](/docs/sdks/capacitor/screen-orientation/) * [ Screen Reader ](/docs/sdks/capacitor/screen-reader/) * [ Screenshot ](/docs/sdks/capacitor/screenshot/) * [ Secure Preferences ](/docs/sdks/capacitor/secure-preferences/) * [ Settings Launcher ](/docs/sdks/capacitor/settings-launcher/) * [ Shake ](/docs/sdks/capacitor/shake/) * [ Silent Mode ](/docs/sdks/capacitor/silent-mode/) * [ SIM ](/docs/sdks/capacitor/sim/) * [ SMS Composer ](/docs/sdks/capacitor/sms-composer/) * [ Speech Recognition ](/docs/sdks/capacitor/speech-recognition/) * [ Speech Synthesis ](/docs/sdks/capacitor/speech-synthesis/) * [ Share Target ](/docs/sdks/capacitor/share-target/) * [ Square Mobile Payments ](/docs/sdks/capacitor/square-mobile-payments/) * [ SQLite ](/docs/sdks/capacitor/sqlite/) * [ Superwall ](/docs/sdks/capacitor/superwall/) * [ System WebView ](/docs/sdks/capacitor/system-webview/) * [ Tauri ](/docs/sdks/capacitor/tauri/) * [ Text Interaction ](/docs/sdks/capacitor/text-interaction/) * [ Text Zoom ](/docs/sdks/capacitor/text-zoom/) * [ Thermal State ](/docs/sdks/capacitor/thermal-state/) * [ Toast ](/docs/sdks/capacitor/toast/) * [ Torch ](/docs/sdks/capacitor/torch/) * [ Vault ](/docs/sdks/capacitor/vault/) * [ Volume ](/docs/sdks/capacitor/volume/) * [ Wallet ](/docs/sdks/capacitor/wallet/) * [ Wifi ](/docs/sdks/capacitor/wifi/) * [ YouTube Player ](/docs/sdks/capacitor/youtube-player/) * [ Zip ](/docs/sdks/capacitor/zip/) * [ Cordova ](/docs/sdks/cordova/) * [ Cloud ](/docs/cloud/) * [ Integrations ](/docs/cloud/live-updates/integrations/) * Concepts * Reference * [ Troubleshooting ](/docs/cloud/live-updates/troubleshooting/) * [ FAQ ](/docs/cloud/live-updates/faq/) * [ Native Builds ](/docs/cloud/native-builds/) * [ Set Up Environments ](/docs/cloud/native-builds/environments/) * [ Overwrite Native Configurations ](/docs/cloud/native-builds/native-configurations/) * [ Auto-Increment Build Numbers ](/docs/cloud/native-builds/auto-incrementing-build-numbers/) * [ Configure the Web Build Script ](/docs/cloud/native-builds/web-build-script/) * [ Build from a Monorepo ](/docs/cloud/native-builds/monorepo/) * [ Use pnpm, Yarn, or bun ](/docs/cloud/native-builds/package-managers/) * [ Install Private npm Packages ](/docs/cloud/native-builds/npm-private-registry/) * [ Override the Java Version ](/docs/cloud/native-builds/override-java-version/) * [ Custom iOS Provisioning Profiles ](/docs/cloud/native-builds/custom-ios-provisioning-profiles/) * [ Build without Git ](/docs/cloud/native-builds/build-without-git/) * [ Access Git Behind a Firewall ](/docs/cloud/native-builds/firewall-access/) * [ Integrations ](/docs/cloud/native-builds/integrations/) * Reference * [ Troubleshooting ](/docs/cloud/native-builds/troubleshooting/) * [ FAQ ](/docs/cloud/native-builds/faq/) * [ App Store Publishing ](/docs/cloud/app-store-publishing/) * [ Submit a Build ](/docs/cloud/app-store-publishing/submit-a-build/) * [ Submit Automatically After a Build ](/docs/cloud/app-store-publishing/submit-automatically/) * [ Troubleshooting ](/docs/cloud/app-store-publishing/troubleshooting/) * [ FAQ ](/docs/cloud/app-store-publishing/faq/) * [ Automations ](/docs/cloud/automations/) * [ Reference ](/docs/cloud/automations/reference/) * [ Troubleshooting ](/docs/cloud/automations/troubleshooting/) * [ FAQ ](/docs/cloud/automations/faq/) * [ Assist ](/docs/cloud/assist/) * [ CLI ](/docs/cloud/cli/) * APIs and SDKs * [ Webhooks ](/docs/cloud/webhooks/) * [ Integrations ](/docs/cloud/integrations/) * Account * [ Organization ](/docs/cloud/organizations/) * [ Two-Factor Enforcement ](/docs/cloud/organizations/two-factor-authentication/) * [ Audit Logs ](/docs/cloud/organizations/audit-logs/) * [ Billing ](/docs/cloud/organizations/billing/) * [ License Keys ](/docs/cloud/license-keys/) * [ AI ](/docs/ai/) * [ Insiders ](/docs/insiders/) * [ Billing & Plans ](/docs/insiders/billing-and-plans/) * [ FAQ ](/docs/insiders/faq/) * [ License ](https://capawesome.io/legal/eula/) * [ Support ](/docs/support/) * [ Contributing ](/docs/contributing/) * Contributing code * [ Code of Conduct ](/docs/contributing/code-of-conduct/) * [ Questions ](https://docs.github.com/en/discussions/collaborating-with-your-community-using-discussions/participating-in-a-discussion#creating-a-discussion) * [ Blog ](/blog/) * Categories # Capacitor ML Kit 8.2.0: 14 New Plugins[¶](#capacitor-ml-kit-820-14-new-plugins "Permanent link") [Capacitor ML Kit](/docs/sdks/capacitor/mlkit/) 8.2.0 is the biggest release in the project's history: 14 new plugins land at once, completing coverage of Google's [ML Kit](https://developers.google.com/ml-kit) APIs. There are new vision plugins for text recognition, image labeling, object detection, pose detection, and digital ink recognition; natural language plugins for smart replies, entity extraction, and language identification; and six GenAI plugins that run Gemini Nano directly on the device. All of it is on-device machine learning: no server, no API keys, and no per-request costs. Here's what's in the release and how to use it. [ ![Build and deploy your Capacitor app with Capawesome Cloud](https://capawesome.io/assets/banners/cloud-build-and-deploy-capacitor-apps.png?t=1) ](https://capawesome.io/) ## At a Glance[¶](#at-a-glance "Permanent link") All 14 plugins were added in [#340](https://github.com/capawesome-team/capacitor-mlkit/pull/340) and published as version 8.2.0 in [#341](https://github.com/capawesome-team/capacitor-mlkit/pull/341): | Plugin | Package | Status | Platforms | | -------------------------------------------------------------------------------- | ----------------------------------------- | ------ | ------------ | | [Image Labeling](/docs/sdks/capacitor/mlkit/image-labeling/) | @capacitor-mlkit/image-labeling | GA | Android, iOS | | [Object Detection](/docs/sdks/capacitor/mlkit/object-detection/) | @capacitor-mlkit/object-detection | GA | Android, iOS | | [Text Recognition](/docs/sdks/capacitor/mlkit/text-recognition/) | @capacitor-mlkit/text-recognition | GA | Android, iOS | | [Pose Detection](/docs/sdks/capacitor/mlkit/pose-detection/) | @capacitor-mlkit/pose-detection | Beta | Android, iOS | | [Digital Ink Recognition](/docs/sdks/capacitor/mlkit/digital-ink-recognition/) | @capacitor-mlkit/digital-ink-recognition | GA | Android, iOS | | [Smart Reply](/docs/sdks/capacitor/mlkit/smart-reply/) | @capacitor-mlkit/smart-reply | GA | Android, iOS | | [Entity Extraction](/docs/sdks/capacitor/mlkit/entity-extraction/) | @capacitor-mlkit/entity-extraction | Beta | Android, iOS | | [Language Identification](/docs/sdks/capacitor/mlkit/language-identification/) | @capacitor-mlkit/language-identification | GA | Android, iOS | | [GenAI Summarization](/docs/sdks/capacitor/mlkit/genai-summarization/) | @capacitor-mlkit/genai-summarization | Beta | Android | | [GenAI Proofreading](/docs/sdks/capacitor/mlkit/genai-proofreading/) | @capacitor-mlkit/genai-proofreading | Beta | Android | | [GenAI Rewriting](/docs/sdks/capacitor/mlkit/genai-rewriting/) | @capacitor-mlkit/genai-rewriting | Beta | Android | | [GenAI Image Description](/docs/sdks/capacitor/mlkit/genai-image-description/) | @capacitor-mlkit/genai-image-description | Beta | Android | | [GenAI Prompt](/docs/sdks/capacitor/mlkit/genai-prompt/) | @capacitor-mlkit/genai-prompt | Beta | Android | | [GenAI Speech Recognition](/docs/sdks/capacitor/mlkit/genai-speech-recognition/) | @capacitor-mlkit/genai-speech-recognition | Alpha | Android | ## Vision Plugins[¶](#vision-plugins "Permanent link") Five new plugins process images on the device, each with a single entry point that takes a local image path and returns structured results. The [Capacitor ML Kit Text Recognition plugin](/docs/sdks/capacitor/mlkit/text-recognition/) extracts text from images using ML Kit Text Recognition v2\. All five script models are bundled with your app (Latin, Chinese, Devanagari, Japanese, Korean), so recognition works offline without any model download. A single call to [processImage(...)](/docs/sdks/capacitor/mlkit/text-recognition/#processimage) returns the full text plus a hierarchy of blocks, lines, and elements, each with bounding boxes and corner points: `[](#%5F%5Fcodelineno-0-1)import { Script, TextRecognition } from '@capacitor-mlkit/text-recognition'; [](#%5F%5Fcodelineno-0-2) [](#%5F%5Fcodelineno-0-3)const processImage = async () => { [](#%5F%5Fcodelineno-0-4) const { text, blocks } = await TextRecognition.processImage({ [](#%5F%5Fcodelineno-0-5) path: 'file:///path/to/image.jpg', [](#%5F%5Fcodelineno-0-6) script: Script.Latin, [](#%5F%5Fcodelineno-0-7) }); [](#%5F%5Fcodelineno-0-8) return { text, blocks }; [](#%5F%5Fcodelineno-0-9)}; ` The other four vision plugins follow the same pattern: * The [Capacitor ML Kit Image Labeling plugin](/docs/sdks/capacitor/mlkit/image-labeling/) identifies objects, places, and activities in an image. Its [processImage(...)](/docs/sdks/capacitor/mlkit/image-labeling/#processimage) method returns labels from a bundled base model covering more than 400 categories, each with a confidence score. * The [Capacitor ML Kit Object Detection plugin](/docs/sdks/capacitor/mlkit/object-detection/) locates objects in an image and returns their bounding boxes via [processImage(...)](/docs/sdks/capacitor/mlkit/object-detection/#processimage). It can detect up to five objects at once and optionally classify each into coarse categories like fashion goods, food, or plants. * The [Capacitor ML Kit Pose Detection plugin](/docs/sdks/capacitor/mlkit/pose-detection/) detects the most prominent person in an image and returns all 33 skeletal landmarks through [processImage(...)](/docs/sdks/capacitor/mlkit/pose-detection/#processimage). You choose between a base model (faster) and an accurate model (better precision) at runtime via the `performanceMode` option. The underlying ML Kit API is in beta. * The [Capacitor ML Kit Digital Ink Recognition plugin](/docs/sdks/capacitor/mlkit/digital-ink-recognition/) turns handwritten strokes into text with [recognize(...)](/docs/sdks/capacitor/mlkit/digital-ink-recognition/#recognize). It supports over 300 languages, plus special models for sketched shapes and emojis. Language models are downloaded on demand with [downloadModel(...)](/docs/sdks/capacitor/mlkit/digital-ink-recognition/#downloadmodel). ## Natural Language Plugins[¶](#natural-language-plugins "Permanent link") Three new plugins analyze text on the device. The [Capacitor ML Kit Smart Reply plugin](/docs/sdks/capacitor/mlkit/smart-reply/) generates up to three contextually relevant reply suggestions for a conversation, the same feature you know from Gmail and Google Messages. You pass the recent messages (up to the last 10 are considered) to [suggestReplies(...)](/docs/sdks/capacitor/mlkit/smart-reply/#suggestreplies) and render the suggestions as quick-reply chips: `[](#%5F%5Fcodelineno-1-1)import { [](#%5F%5Fcodelineno-1-2) SmartReply, [](#%5F%5Fcodelineno-1-3) SmartReplySuggestionResultStatus, [](#%5F%5Fcodelineno-1-4)} from '@capacitor-mlkit/smart-reply'; [](#%5F%5Fcodelineno-1-5) [](#%5F%5Fcodelineno-1-6)const suggestReplies = async () => { [](#%5F%5Fcodelineno-1-7) const { status, suggestions } = await SmartReply.suggestReplies({ [](#%5F%5Fcodelineno-1-8) messages: [ [](#%5F%5Fcodelineno-1-9) { [](#%5F%5Fcodelineno-1-10) text: 'Are you free for lunch today?', [](#%5F%5Fcodelineno-1-11) timestamp: Date.now() - 60_000, [](#%5F%5Fcodelineno-1-12) isLocalUser: false, [](#%5F%5Fcodelineno-1-13) userId: 'user-1', [](#%5F%5Fcodelineno-1-14) }, [](#%5F%5Fcodelineno-1-15) { [](#%5F%5Fcodelineno-1-16) text: 'Sure, where do you want to go?', [](#%5F%5Fcodelineno-1-17) timestamp: Date.now() - 30_000, [](#%5F%5Fcodelineno-1-18) isLocalUser: true, [](#%5F%5Fcodelineno-1-19) }, [](#%5F%5Fcodelineno-1-20) { [](#%5F%5Fcodelineno-1-21) text: 'How about the new place downtown?', [](#%5F%5Fcodelineno-1-22) timestamp: Date.now(), [](#%5F%5Fcodelineno-1-23) isLocalUser: false, [](#%5F%5Fcodelineno-1-24) userId: 'user-1', [](#%5F%5Fcodelineno-1-25) }, [](#%5F%5Fcodelineno-1-26) ], [](#%5F%5Fcodelineno-1-27) }); [](#%5F%5Fcodelineno-1-28) [](#%5F%5Fcodelineno-1-29) if (status === SmartReplySuggestionResultStatus.Success) { [](#%5F%5Fcodelineno-1-30) return suggestions; [](#%5F%5Fcodelineno-1-31) } [](#%5F%5Fcodelineno-1-32) return []; [](#%5F%5Fcodelineno-1-33)}; ` Note that Smart Reply currently supports English conversations only; for other languages, the result status tells you no suggestions are available. The two remaining plugins cover text analysis: * The [Capacitor ML Kit Entity Extraction plugin](/docs/sdks/capacitor/mlkit/entity-extraction/) recognizes 11 types of structured entities in raw text: addresses, date-times, email addresses, flight numbers, IBANs, ISBNs, money amounts, payment card numbers, phone numbers, tracking numbers, and URLs. Call [extractEntities(...)](/docs/sdks/capacitor/mlkit/entity-extraction/#extractentities) after downloading the model for one of the 15 supported languages. The underlying ML Kit API is in beta. * The [Capacitor ML Kit Language Identification plugin](/docs/sdks/capacitor/mlkit/language-identification/) detects the language of a string among more than 100 languages, returned as BCP-47 tags. [identifyLanguage(...)](/docs/sdks/capacitor/mlkit/language-identification/#identifylanguage) returns the most likely language, while [identifyPossibleLanguages(...)](/docs/sdks/capacitor/mlkit/language-identification/#identifypossiblelanguages) returns all candidates with confidence scores. No model download is required. ## On-Device GenAI with Gemini Nano[¶](#on-device-genai-with-gemini-nano "Permanent link") The headline of this release: six plugins bring Google's [ML Kit GenAI APIs](https://developers.google.com/ml-kit/genai) to Capacitor. They run Gemini Nano, Google's most efficient foundation model, directly on the device via Android's AICore. Prompts and results never leave the phone, inference works offline, and there is no API key to manage and no per-request billing. The trade-off is hardware support: the GenAI plugins are Android-only and require a device that supports Gemini Nano (see Google's [list of supported devices](https://developers.google.com/ml-kit/genai)) with Android API level 26 or higher. On iOS and the web, all methods reject with an unavailability error, so you can feature-detect and fall back gracefully. All six plugins share the same lifecycle. You first check whether the feature is available on the device, then download the model weights if needed, then run inference. Here's what that looks like with the [Capacitor ML Kit GenAI Summarization plugin](/docs/sdks/capacitor/mlkit/genai-summarization/), using [checkFeatureStatus(...)](/docs/sdks/capacitor/mlkit/genai-summarization/#checkfeaturestatus) and [downloadFeature(...)](/docs/sdks/capacitor/mlkit/genai-summarization/#downloadfeature) with a `downloadProgress` listener: `[](#%5F%5Fcodelineno-2-1)import { [](#%5F%5Fcodelineno-2-2) FeatureStatus, [](#%5F%5Fcodelineno-2-3) GenAiSummarization, [](#%5F%5Fcodelineno-2-4) InputType, [](#%5F%5Fcodelineno-2-5) Language, [](#%5F%5Fcodelineno-2-6) OutputType, [](#%5F%5Fcodelineno-2-7)} from '@capacitor-mlkit/genai-summarization'; [](#%5F%5Fcodelineno-2-8) [](#%5F%5Fcodelineno-2-9)const prepareFeature = async () => { [](#%5F%5Fcodelineno-2-10) const options = { [](#%5F%5Fcodelineno-2-11) inputType: InputType.Article, [](#%5F%5Fcodelineno-2-12) language: Language.English, [](#%5F%5Fcodelineno-2-13) outputType: OutputType.ThreeBullets, [](#%5F%5Fcodelineno-2-14) }; [](#%5F%5Fcodelineno-2-15) const { featureStatus } = await GenAiSummarization.checkFeatureStatus(options); [](#%5F%5Fcodelineno-2-16) if (featureStatus === FeatureStatus.Unavailable) { [](#%5F%5Fcodelineno-2-17) return false; [](#%5F%5Fcodelineno-2-18) } [](#%5F%5Fcodelineno-2-19) if (featureStatus === FeatureStatus.Downloadable) { [](#%5F%5Fcodelineno-2-20) await GenAiSummarization.addListener('downloadProgress', (event) => { [](#%5F%5Fcodelineno-2-21) console.log('Total bytes downloaded:', event.totalBytesDownloaded); [](#%5F%5Fcodelineno-2-22) }); [](#%5F%5Fcodelineno-2-23) await GenAiSummarization.downloadFeature(options); [](#%5F%5Fcodelineno-2-24) } [](#%5F%5Fcodelineno-2-25) return true; [](#%5F%5Fcodelineno-2-26)}; ` Once the feature is ready, you can summarize an article or conversation with [summarize(...)](/docs/sdks/capacitor/mlkit/genai-summarization/#summarize). Inference streams: the `inferenceProgress` listener delivers partial results as they are generated, so you can render the output token by token like a chat UI, while the returned promise resolves with the complete summary: `[](#%5F%5Fcodelineno-3-1)const summarize = async (text: string) => { [](#%5F%5Fcodelineno-3-2) await GenAiSummarization.addListener('inferenceProgress', (event) => { [](#%5F%5Fcodelineno-3-3) console.log('Partial result:', event.text); [](#%5F%5Fcodelineno-3-4) }); [](#%5F%5Fcodelineno-3-5) const { summary } = await GenAiSummarization.summarize({ [](#%5F%5Fcodelineno-3-6) text, [](#%5F%5Fcodelineno-3-7) inputType: InputType.Article, [](#%5F%5Fcodelineno-3-8) language: Language.English, [](#%5F%5Fcodelineno-3-9) outputType: OutputType.ThreeBullets, [](#%5F%5Fcodelineno-3-10) }); [](#%5F%5Fcodelineno-3-11) return summary; [](#%5F%5Fcodelineno-3-12)}; ` The same `checkFeatureStatus` → `downloadFeature` → inference flow, including the streaming `inferenceProgress` events, applies to the other GenAI plugins: * The [Capacitor ML Kit GenAI Proofreading plugin](/docs/sdks/capacitor/mlkit/genai-proofreading/) fixes grammar and spelling with [proofread(...)](/docs/sdks/capacitor/mlkit/genai-proofreading/#proofread), returning suggestions sorted by confidence. It supports English, French, German, Italian, Japanese, Korean, and Spanish. * The [Capacitor ML Kit GenAI Rewriting plugin](/docs/sdks/capacitor/mlkit/genai-rewriting/) rephrases text in a different tone via [rewrite(...)](/docs/sdks/capacitor/mlkit/genai-rewriting/#rewrite). Output types include `ELABORATE`, `EMOJIFY`, `SHORTEN`, `FRIENDLY`, `PROFESSIONAL`, and `REPHRASE`. * The [Capacitor ML Kit GenAI Image Description plugin](/docs/sdks/capacitor/mlkit/genai-image-description/) generates a short description of an image with [describeImage(...)](/docs/sdks/capacitor/mlkit/genai-image-description/#describeimage), useful for accessibility labels and media metadata. Descriptions are currently English only. * The [Capacitor ML Kit GenAI Prompt plugin](/docs/sdks/capacitor/mlkit/genai-prompt/) is the open-ended one: [generateContent(...)](/docs/sdks/capacitor/mlkit/genai-prompt/#generatecontent) sends a custom prompt (optionally with an image) to Gemini Nano, with tuning parameters like `temperature`, `topK`, and `maxOutputTokens`. Prompts are limited to about 4,000 tokens, and unlike the task-specific APIs, the output is not moderated, so it's your responsibility to use it carefully. * The [Capacitor ML Kit GenAI Speech Recognition plugin](/docs/sdks/capacitor/mlkit/genai-speech-recognition/) transcribes speech on-device in a session-based flow: [startRecognition(...)](/docs/sdks/capacitor/mlkit/genai-speech-recognition/#startrecognition) begins listening and emits `partialResult` and `finalResult` events until you call [stopRecognition()](/docs/sdks/capacitor/mlkit/genai-speech-recognition/#stoprecognition). A word of caution: this one is in **alpha**. Its basic mode requires Android API level 31+, its advanced GenAI mode currently runs on a very limited set of devices (such as the Pixel 10), and breaking changes are expected. ## Getting Started with the New Capacitor ML Kit Plugins[¶](#getting-started-with-the-new-capacitor-ml-kit-plugins "Permanent link") Every plugin ships as its own npm package, so you only install what you use, and each one supports Capacitor 8\. The cross-platform plugins require an iOS deployment target of 15.5 or higher and are currently CocoaPods-only. For setup instructions, refer to the Installation section in each plugin's documentation, for example [Text Recognition](/docs/sdks/capacitor/mlkit/text-recognition/#installation) or [GenAI Summarization](/docs/sdks/capacitor/mlkit/genai-summarization/#installation). The [Capacitor ML Kit overview](/docs/sdks/capacitor/mlkit/) lists all 21 plugins in the project. To get notified when the next batch of plugins and features ships, subscribe to our newsletter: [Subscribe to the Capawesome Newsletter](https://capawesome.io/newsletter) ## Final Thoughts[¶](#final-thoughts "Permanent link") If you've been calling a cloud API for tasks like text recognition or summarization, this release is a good moment to reconsider: on-device inference removes the network round trip, works in airplane mode, and keeps user data on the phone. A practical starting point is one of the GA vision plugins, which need no model download and work on both platforms. Treat the GenAI plugins as an early look — they are in beta (and Speech Recognition in alpha), and device support for Gemini Nano is still expanding. For some history on the project, read the announcement of its very first plugin: [Announcing the Capacitor ML Kit Barcode Scanning Plugin](/blog/announcing-the-capacitor-mlkit-barcode-scanner-plugin/). Got questions or feedback on the new plugins? Join the [Capawesome Discord server](https://discord.gg/VCXxSVjefW) and let us know what you're building. And to stay up to date on future releases, subscribe to the [Capawesome newsletter](https://capawesome.io/newsletter). July 20, 2026 Back to top ```json { "@context": "https://schema.org", "@type": "BlogPosting", "headline": "Capacitor ML Kit 8.2.0: 14 New Plugins", "description": "Capacitor ML Kit 8.2.0 adds 14 new plugins for on-device machine learning, including six GenAI plugins powered by Gemini Nano on Android.", "image": "https://capawesome.io/assets/banners/cloud-build-and-deploy-capacitor-apps.png", "datePublished": "2026-07-20T00:00:00+00:00", "dateModified": "2026-07-20T00:00:00+00:00", "author": [ { "@type": "Person", "name": "Robin Genz", "url": "https://github.com/robingenz" } ], "publisher": { "@type": "Organization", "name": "Capawesome", "url": "https://capawesome.io", "logo": { "@type": "ImageObject", "url": "https://capawesome.io/assets/images/logo.svg" } }, "articleSection": "Announcements", "keywords": ["Announcements", "Capacitor", "ML Kit", "SDKs"], "isPartOf": { "@type": "Blog", "@id": "https://capawesome.io/blog/#blog" }, "mainEntityOfPage": "https://capawesome.io/blog/capacitor-mlkit-8-2-0-release/", "url": "https://capawesome.io/blog/capacitor-mlkit-8-2-0-release/" } { "@context": "https://schema.org", "@type": "BreadcrumbList", "itemListElement": [ { "@type": "ListItem", "position": 1, "name": "Home", "item": "https://capawesome.io/" }, { "@type": "ListItem", "position": 2, "name": "Blog", "item": "https://capawesome.io/blog/" }, { "@type": "ListItem", "position": 3, "name": "Capacitor ML Kit 8.2.0: 14 New Plugins", "item": "https://capawesome.io/blog/capacitor-mlkit-8-2-0-release/" } ] } ```