[[["易于理解","easyToUnderstand","thumb-up"],["解决了我的问题","solvedMyProblem","thumb-up"],["其他","otherUp","thumb-up"]],[["很难理解","hardToUnderstand","thumb-down"],["信息或示例代码不正确","incorrectInformationOrSampleCode","thumb-down"],["没有我需要的信息/示例","missingTheInformationSamplesINeed","thumb-down"],["翻译问题","translationIssue","thumb-down"],["其他","otherDown","thumb-down"]],["最后更新时间 (UTC):2025-04-03。"],[],[],null,["# What is Cloud Translation?\n==========================\n\nCloud Translation enables your websites and applications to dynamically translate\ntext **programmatically** through an API. Cloud Translation uses a Google\npre-trained or a custom machine learning model to translate text. By\ndefault, Cloud Translation uses the Google pre-trained\nNeural Machine Translation (NMT) model, which Google updates on\na semi-regular cadence with the availability of more training data or better\ntechniques.\n\nWith Cloud Translation, you can also perform translations through the\nTranslation LLM which is fine-tuned from Gemini foundation models. This\nmodel offers higher quality, more natural sounding translations and performs\nespecially well translating idioms and sayings compared to other models.\n| **Note:** No customer data and translations are used to improve the models. For more information about data usage, refer to the [Data usage FAQ](/translate/data-usage).\n\nBenefits\n--------\n\nCloud Translation can translate text for more than 100 language pairs. If\nyou don't know the language of your source text, Cloud Translation can\ndetect it for you. For a list of all of the supported languages, see [Supported\nlanguages](/translate/docs/languages).\n\nCloud Translation scales seamlessly and allows unlimited character translations\nper day. However, there are restrictions on content size for each request and\nrequest rates. Additionally, you can use quota limits to manage your budget. For\nmore information, see [Quotas and limits](/translate/quotas).\n\n### Example use case\n\nAs an enterprise organization, you can include translated text in existing\ninternal workflows. For example, you might have a global support team that\nreceives customer cases in many different languages. You can add\nCloud Translation into your workflow to translate those customer cases into\nthe native language of your support team members.\n\nBy using Cloud Translation, you improve the overall efficiency of your\nsupport team because they don't need to manually translate text themselves or\nrely on others.\n\nCloud Translation editions\n--------------------------\n\nCloud Translation offers two editions:\n\n- Cloud Translation - Basic (v2)\n- Cloud Translation - Advanced (v3)\n\nBoth editions support language detection and translation. They also use the same\nGoogle pre-trained model to translate content. Cloud Translation - Advanced\nincludes features such as batch requests, AutoML custom models,\nand glossaries. For example, if you want to consistently translate terms that\nare specific to your use case, you can use glossaries to do that.\n\nThe edition you use depends on which version of the client libraries or service\nendpoint that you use.\n\n\nIf you are planning a new project, build your application with\nCloud Translation - Advanced to take advantage of new features and service\nimprovements. Cloud Translation - Basic remains available, but does not support new\nfeatures like using glossaries, requesting batch translations, or using\nAutoML models.\n\n\u003cbr /\u003e\n\nFor a comparison between the two editions, see [Comparing Basic and\nAdvanced](/translate/docs/editions).\n\n### Migrate to Advanced from Basic\n\n\nIf already have existing applications that use Cloud Translation - Basic, you can\nmigrate them to use Cloud Translation - Advanced. To use Cloud Translation - Advanced,\nyou must update applications to use, for example, service accounts and the\nAdvanced edition client libraries. For more information, see [Migrating to\nCloud Translation - Advanced](/translate/docs/migrate-to-v3).\n\n\u003cbr /\u003e\n\nCustom translations (Advanced only)\n-----------------------------------\n\nIn cases where the default model isn't generating high-quality results for a\nparticular linguistic domain, you can provide a set of example translations to\ngenerate custom translations. Cloud Translation uses your examples to\ngenerate translations that follow your style, tone, and vocabulary. For example,\ncustom translations might be useful if you regularly translate industry-specific\ncontent from the medical field or marketing content that needs to follow a\nspecific style.\n\nCloud Translation provides two methods for customizing translations. You\ncan train a custom model through AutoML Translation, orcustom- you can use\nadaptive translation. Train a model with\nAutoML Translation if your content uses a lot of industry-specific\nterminology. Adaptive translation might be a better fit if translated content\nshould follow a unique style. For more information, see the [Custom\ntranslations overview](/translate/docs/advanced/custom-translations).\n\nPricing\n-------\n\nCloud Translation charges you on a monthly basis based on the number of\ncharacters that you send. For more information, see\n[Pricing](/translate/pricing).\n\nWhat's next\n-----------\n\n- To start using the API, get [set up](/translate/docs/setup) and then try the [Basic quickstart guide](/translate/docs/basic/translate-text-basic) or the [Advanced quickstart guide](/translate/docs/advanced/translate-text-advanced).\n- For answers to general questions about Cloud Translation, refer to the [General FAQs](/translate/faq) page."]]