机器学习套件是一个移动 SDK，可将 Google 的设备端机器学习专业技术融入到 Android 和 iOS 应用中。使用功能强大且易于使用的 GenAI、Vision 和 Natural Language API，解决应用中的常见问题或打造全新的用户体验。所有这些功能均由 Google 一流的机器学习模型提供支持，并且免费提供给您。
机器学习套件的 API 全部在设备端运行，因此可支持需要处理实时相机数据流的实时用例。这也意味着，您可以在线下使用此功能。
[null,null,["最后更新时间 (UTC)：2025-08-29。"],[[["\u003cp\u003eML Kit is a free mobile SDK that brings Google's machine learning expertise to Android and iOS apps through easy-to-use Vision and Natural Language APIs.\u003c/p\u003e\n"],["\u003cp\u003eML Kit's APIs run on-device, enabling real-time processing and offline functionality, but require Android API level 21 or higher.\u003c/p\u003e\n"],["\u003cp\u003eDevelopers can utilize pre-built APIs for tasks like text recognition, image labeling, and language identification, or integrate custom TensorFlow Lite models.\u003c/p\u003e\n"],["\u003cp\u003eGoogle provides extensive resources including sample apps, codelabs, and documentation to help developers get started with ML Kit.\u003c/p\u003e\n"],["\u003cp\u003eWhile most of ML Kit is generally available, certain features like Pose Detection and Text Recognition v2 are currently in beta.\u003c/p\u003e\n"]]],[],null,["Use on-device machine learning in your apps to easily solve real-world problems.\n\nML Kit is a mobile SDK that brings Google's on-device machine learning\nexpertise to Android and iOS apps. Use our powerful yet easy to use GenAI,\nVision and Natural Language APIs to solve common challenges in your apps or\ncreate brand-new user experiences. All are powered by Google's best-in-class ML\nmodels and offered to you at no cost.\n\nML Kit's APIs all run on-device, allowing for real-time use cases where you want\nto process a live camera stream for example. This also means that the\nfunctionality is available offline.\n| All ML Kit APIs require Android API level 21 or higher.\n\nWhat's new \n\n- We launched the beta version of the [GenAI APIs](/ml-kit/genai), offering out-of-the-box quality for common use cases through a high-level interface. Initial launch includes [Image Description](/ml-kit/genai/image-description/android), [Proofreading](/ml-kit/genai/proofreading/android), [Rewriting](/ml-kit/genai/rewriting/android) and [Summarization](/ml-kit/genai/summarization/android).\n\nLearn more\n\n- [GenAI APIs](/ml-kit/genai) harness the power of Gemini Nano to help your app perform tasks. These APIs provide out-of-the-box quality for popular use cases such as: [Image Description](/ml-kit/genai/image-description/android), [Proofreading](/ml-kit/genai/proofreading/android), [Rewriting](/ml-kit/genai/rewriting/android) and [Summarization](/ml-kit/genai/summarization/android).\n- ML Kit also offers ready-to-use APIs powered by Google's ML models: [text recognition](/ml-kit/vision/text-recognition), [face detection](/ml-kit/vision/face-detection), [barcode scanning](/ml-kit/vision/barcode-scanning), [image labeling](/ml-kit/vision/image-labeling), [object detection and tracking](/ml-kit/vision/object-detection), [pose detection](/ml-kit/vision/pose-detection), [selfie segmentation](/ml-kit/vision/selfie-segmentation), [smart reply](/ml-kit/language/smart-reply), [text translation](/ml-kit/language/translation), and [language identification](/ml-kit/language/identification).\n- Learn how to use custom TensorFlow Lite image labeling models in your apps. Read [Custom models with ML Kit](/ml-kit/custom-models).\n- Take a look at our [sample apps and codelabs](/ml-kit/samples). They help you get started with all of the APIs.\n\nOther resources\n\nIf ML Kit's turn-key APIs don't meet your needs and you require a more custom\nsolution, visit the\n[On-device Machine Learning page](/learn/topics/on-device-ml) for guidance on\nall of Google's solutions and tools for on-device machine learning."]]
