AWS bedrock documentation change
Summary
Translated documentation from Chinese to English, updated section titles, links, and content to English equivalents. Added PDF link reference and timeout guidance for Claude 3.7/4 models.
Security assessment
Changes are primarily language translation and content reorganization. The timeout guidance for SDK clients (60 minutes) is a best practice for long-running operations but doesn't address a specific vulnerability. No security vulnerabilities, exploits, or security features are mentioned in the diff.
Diff
diff --git a/bedrock/latest/userguide/model-parameters-anthropic-claude-messages.md b/bedrock/latest/userguide/model-parameters-anthropic-claude-messages.md index af3b0d55f..399321f87 100644 --- a//bedrock/latest/userguide/model-parameters-anthropic-claude-messages.md +++ b//bedrock/latest/userguide/model-parameters-anthropic-claude-messages.md @@ -1 +1 @@ -[文档](/index.html)[Amazon Bedrock](/bedrock/index.html)[用户指南](what-is-bedrock.html) +[](/pdfs/bedrock/latest/userguide/bedrock-ug.pdf#model-parameters-anthropic-claude-messages "Open PDF") @@ -3 +3 @@ -AnthropicClaude Messages API 概述 +[Documentation](/index.html)[Amazon Bedrock](/bedrock/index.html)[User Guide](what-is-bedrock.html) @@ -5 +5 @@ AnthropicClaude Messages API 概述 -本文属于机器翻译版本。若本译文内容与英语原文存在差异，则一律以英文原文为准。 +Anthropic Claude Messages API overview @@ -9 +9 @@ AnthropicClaude Messages API 概述 -本部分介绍使用 Anthropic Claude Messages API 所需的推理参数和代码示例。 +This section provides inference parameters and code examples for using the Anthropic Claude Messages API. @@ -11 +11 @@ AnthropicClaude Messages API 概述 -###### 主题 +###### Topics @@ -13 +13 @@ AnthropicClaude Messages API 概述 - * AnthropicClaude Messages API 概述 + * Anthropic Claude Messages API overview @@ -15 +15 @@ AnthropicClaude Messages API 概述 - * [工具使用](./model-parameters-anthropic-claude-messages-tool-use.html) + * [Tool use](./model-parameters-anthropic-claude-messages-tool-use.html) @@ -17 +17 @@ AnthropicClaude Messages API 概述 - * [延伸思考](./claude-messages-extended-thinking.html) + * [Extended thinking](./claude-messages-extended-thinking.html) @@ -19 +19 @@ AnthropicClaude Messages API 概述 - * [思考加密](./claude-messages-thinking-encryption.html) + * [Thinking encryption](./claude-messages-thinking-encryption.html) @@ -21 +21 @@ AnthropicClaude Messages API 概述 - * [不同模型版本在思考上的不同](./claude-messages-thinking-differences.html) + * [Differences in thinking across model versions](./claude-messages-thinking-differences.html) @@ -23 +23 @@ AnthropicClaude Messages API 概述 - * [请求和响应](./model-parameters-anthropic-claude-messages-request-response.html) + * [Request and Response](./model-parameters-anthropic-claude-messages-request-response.html) @@ -25 +25 @@ AnthropicClaude Messages API 概述 - * [代码示例](./api-inference-examples-claude-messages-code-examples.html) + * [Code examples](./api-inference-examples-claude-messages-code-examples.html) @@ -27 +27 @@ AnthropicClaude Messages API 概述 - * [支持的模型](./claude-messages-supported-models.html) + * [Supported models](./claude-messages-supported-models.html) @@ -32 +32 @@ AnthropicClaude Messages API 概述 -## AnthropicClaude Messages API 概述 +## Anthropic Claude Messages API overview @@ -34 +34 @@ AnthropicClaude Messages API 概述 -您可以使用 Messages API 来创建聊天机器人或虚拟助手应用程序。API 管理用户和 Anthropic Claude 模型（助手）之间的对话交流。 +You can use the Messages API to create chat bots or virtual assistant applications. The API manages the conversational exchanges between a user and an Anthropic Claude model (assistant). @@ -36 +36 @@ AnthropicClaude Messages API 概述 -###### 提示 +###### Tip @@ -38 +38 @@ AnthropicClaude Messages API 概述 -本主题介绍如何将AnthropicClaude消息 API 与基本推理操作（[InvokeModel](https://docs.aws.amazon.com/bedrock/latest/APIReference/API_runtime_InvokeModel.html)或 [InvokeModelWithResponseStream](https://docs.aws.amazon.com/bedrock/latest/APIReference/API_runtime_InvokeModelWithResponseStream.html)）配合使用。但是，我们建议您使用 Converse API 在应用程序中实施消息传递。Converse API 提供了一组统一的参数，适用于所有支持消息的模型。有关更多信息，请参阅 [使用 Converse API 操作进行对话](./conversation-inference.html)。 +This topic shows how to use the Anthropic Claude messages API with the base inference operations ([InvokeModel](https://docs.aws.amazon.com/bedrock/latest/APIReference/API_runtime_InvokeModel.html) or [InvokeModelWithResponseStream](https://docs.aws.amazon.com/bedrock/latest/APIReference/API_runtime_InvokeModelWithResponseStream.html)). However, we recommend that you use the Converse API to implement messages in your application. The Converse API provides a unified set of parameters that work across all models that support messages. For more information, see [Carry out a conversation with the Converse API operations](./conversation-inference.html). @@ -40 +40 @@ AnthropicClaude Messages API 概述 -Anthropic 训练 Claude 模型在交替的用户和助手对话回合中进行操作。创建新消息时，您可以使用消息参数指定之前的对话回合。然后，模型在对话中生成下一条消息。 +Anthropic trains Claude models to operate on alternating user and assistant conversational turns. When creating a new message, you specify the prior conversational turns with the messages parameter. The model then generates the next Message in the conversation. @@ -42 +42 @@ Anthropic 训练 Claude 模型在交替的用户和助手对话回合中进行 -每条输入消息都必须是一个具有角色和内容的对象。您可以指定一条用户角色消息，也可以包含多条用户和助手消息。 +Each input message must be an object with a role and content. You can specify a single user-role message, or you can include multiple user and assistant messages. @@ -44 +44 @@ Anthropic 训练 Claude 模型在交替的用户和助手对话回合中进行 -如果您使用的是预填 Claude 响应的技术（使用最后一条助手角色消息填写 Claude 响应的开头），Claude 将会从您上次停下来的地方继续响应。使用这种技术，Claude 仍会返回带有助手角色的响应。 +If you are using the technique of prefilling the response from Claude (filling in the beginning of Claude's response by using a final assistant role Message), Claude will respond by picking up from where you left off. With this technique, Claude will still return a response with the assistant role. @@ -46 +46 @@ Anthropic 训练 Claude 模型在交替的用户和助手对话回合中进行 -如果最后一条消息使用助手角色，系统会立即根据该消息中的内容继续发送响应内容。您可以用它来限制模型的部分响应。 +If the final message uses the assistant role, the response content will continue immediately from the content in that message. You can use this to constrain part of the model's response. @@ -48 +48 @@ Anthropic 训练 Claude 模型在交替的用户和助手对话回合中进行 -单个用户消息示例： +Example with a single user message: @@ -53 +53 @@ Anthropic 训练 Claude 模型在交替的用户和助手对话回合中进行 -多个对话回合示例： +Example with multiple conversational turns: @@ -62 +62 @@ Anthropic 训练 Claude 模型在交替的用户和助手对话回合中进行 -部分填充 Claude 响应示例： +Example with a partially-filled response from Claude: @@ -70 +70 @@ Anthropic 训练 Claude 模型在交替的用户和助手对话回合中进行 -每个输入消息内容可以是单个字符串，也可以是内容块数组，其中每个内容块都有特定的类型。使用字符串是“文本”类型的一个内容块数组的简称。以下输入消息具有同等效果： +Each input message content may be either a single string or an array of content blocks, where each block has a specific type. Using a string is shorthand for an array of one content block of type "text". The following input messages are equivalent: @@ -78 +78 @@ Anthropic 训练 Claude 模型在交替的用户和助手对话回合中进行 -有关为 Anthropic Claude 模型创建提示的信息，请参阅 Anthropic Claude 文档中的[提示简介](https://docs.anthropic.com/claude/docs/intro-to-prompting)。如果您想将现有的[文本补全](./model-parameters-anthropic-claude-text-completion.html)提示迁移到 Messages API，请参阅[从文本补全迁移](https://docs.anthropic.com/claude/reference/migrating-from-text-completions-to-messages)。 +For information about creating prompts for Anthropic Claude models, see [Intro to prompting](https://docs.anthropic.com/claude/docs/intro-to-prompting) in the Anthropic Claude documentation. If you have existing [Text Completion](./model-parameters-anthropic-claude-text-completion.html) prompts that you want to migrate to the messages API, see [Migrating from Text Completions](https://docs.anthropic.com/claude/reference/migrating-from-text-completions-to-messages). @@ -80 +80 @@ Anthropic 训练 Claude 模型在交替的用户和助手对话回合中进行 -###### 重要 +###### Important @@ -82 +82 @@ Anthropic 训练 Claude 模型在交替的用户和助手对话回合中进行 -对 Anthropic Claude 3.7 Sonnet 和 Claude 4 模型进行推理调用的超时时间为 60 分钟。默认情况下，AWSSDK 客户端会在 1 分钟后超时。我们建议您将 AWS SDK 客户端的读取超时时间延长至至少 60 分钟。例如，在 AWS Python botocore SDK 中，请将 [botocore.config](https://botocore.amazonaws.com/v1/documentation/api/latest/reference/config.html#) 中的 `read_timeout` 字段值更改为至少 3600。 +The timeout period for inference calls to Anthropic Claude 3.7 Sonnet and Claude 4 models is 60 minutes. By default, AWS SDK clients timeout after 1 minute. We recommend that you increase the read timeout period of your AWS SDK client to at least 60 minutes. For example, in the AWS Python botocore SDK, change the value of the `read_timeout` field in [botocore.config](https://botocore.amazonaws.com/v1/documentation/api/latest/reference/config.html#) to at least 3600. @@ -84 +84 @@ Anthropic 训练 Claude 模型在交替的用户和助手对话回合中进行 -### 系统提示 +### System prompts @@ -86 +86 @@ Anthropic 训练 Claude 模型在交替的用户和助手对话回合中进行 -您还可以在请求中包含系统提示。系统提示允许您为 Anthropic Claude 提供上下文和说明，例如指定特定的目标或角色。在 `system` 字段中指定系统提示，如以下示例所示。 +You can also include a system prompt in the request. A system prompt lets you provide context and instructions to Anthropic Claude, such as specifying a particular goal or role. Specify a system prompt in the `system` field, as shown in the following example. @@ -93 +93 @@ Anthropic 训练 Claude 模型在交替的用户和助手对话回合中进行 -有关更多信息，请参阅 Anthropic 文档中的[系统提示](https://docs.anthropic.com/en/docs/system-prompts)。 +For more information, see [System prompts](https://docs.anthropic.com/en/docs/system-prompts) in the Anthropic documentation. @@ -95 +95 @@ Anthropic 训练 Claude 模型在交替的用户和助手对话回合中进行 -### 多模态提示 +### Multimodal prompts @@ -97 +97 @@ Anthropic 训练 Claude 模型在交替的用户和助手对话回合中进行 -多模态提示将多种模态（图像和文本）组合到一个提示中。您可以在 `content` 输入字段中指定模态。以下示例显示了如何要求 Anthropic Claude 描述所提供图像的内容。有关示例代码，请参阅 [多模态代码示例](./api-inference-examples-claude-messages-code-examples.html#api-inference-examples-claude-multimodal-code-example)。 +A multimodal prompt combines multiple modalities (images and text) in a single prompt. You specify the modalities in the `content` input field. The following example shows how you could ask Anthropic Claude to describe the content of a supplied image. For example code, see [Multimodal code examples](./api-inference-examples-claude-messages-code-examples.html#api-inference-examples-claude-multimodal-code-example). @@ -124 +124 @@ Anthropic 训练 Claude 模型在交替的用户和助手对话回合中进行 -###### 注意 +###### Note @@ -126 +126 @@ Anthropic 训练 Claude 模型在交替的用户和助手对话回合中进行 -以下限制适用于 `content` 字段： +The following restrictions pertain to the `content` field: @@ -128 +128 @@ Anthropic 训练 Claude 模型在交替的用户和助手对话回合中进行 - * 您最多可以包含 20 个图像。每个图像的大小、高度和宽度必须分别不超过 3.75 MB、8000 像素和 8000 像素。 + * You can include up to 20 images. Each image's size, height, and width must be no more than 3.75 MB, 8,000 px, and 8,000 px, respectively. @@ -130 +130 @@ Anthropic 训练 Claude 模型在交替的用户和助手对话回合中进行 - * 您最多可以包含五个文档。每个文档的大小不得超过 4.5 MB。 + * You can include up to five documents. Each document's size must be no more than 4.5 MB. @@ -132 +132 @@ Anthropic 训练 Claude 模型在交替的用户和助手对话回合中进行 - * 如果 `role` 是 `user`，则只能包含图像和文档。 + * You can only include images and documents if the `role` is `user`. @@ -137 +137 @@ Anthropic 训练 Claude 模型在交替的用户和助手对话回合中进行 -您在请求中包含的每张图像都计入您的词元使用量。有关更多信息，请参阅 Anthropic 文档中的[图像成本](https://docs.anthropic.com/claude/docs/vision#image-costs)。 +Each image you include in a request counts towards your token usage. For more information, see [Image costs](https://docs.anthropic.com/claude/docs/vision#image-costs) in the Anthropic documentation. @@ -139 +139 @@ Anthropic 训练 Claude 模型在交替的用户和助手对话回合中进行 - **Javascript 在您的浏览器中被禁用或不可用。** + **Javascript is disabled or is unavailable in your browser.** @@ -141 +141 @@ Anthropic 训练 Claude 模型在交替的用户和助手对话回合中进行 -要使用 Amazon Web Services 文档，必须启用 Javascript。请参阅浏览器的帮助页面以了解相关说明。 +To use the Amazon Web Services Documentation, Javascript must be enabled. Please refer to your browser's Help pages for instructions. @@ -143 +143 @@ Anthropic 训练 Claude 模型在交替的用户和助手对话回合中进行 -[文档惯例](/general/latest/gr/docconventions.html) +[Document Conventions](/general/latest/gr/docconventions.html) @@ -147 +147 @@ Anthropic Claude Text Completions API -工具使用 +Tool use @@ -149 +149 @@ Anthropic Claude Text Completions API -此页面对您有帮助吗？- 是 +Did this page help you? - Yes @@ -151 +151 @@ Anthropic Claude Text Completions API -感谢您对我们工作的肯定！ +Thanks for letting us know we're doing a good job! @@ -153 +153 @@ Anthropic Claude Text Completions API -如果不耽误您的时间，请告诉我们做得好的地方，让我们做得更好。 +If you've got a moment, please tell us what we did right so we can do more of it. @@ -155 +155 @@ Anthropic Claude Text Completions API -此页面对您有帮助吗？- 否 +Did this page help you? - No @@ -157 +157 @@ Anthropic Claude Text Completions API -感谢您告诉我们本页内容还需要完善。很抱歉让您失望了。 +Thanks for letting us know this page needs work. We're sorry we let you down. @@ -159 +159 @@ Anthropic Claude Text Completions API -如果不耽误您的时间，请告诉我们如何改进文档。 +If you've got a moment, please tell us how we can make the documentation better.