AWS bedrock documentation change
Summary
Complete restructuring of the overview page: replaced conceptual explanations with API code examples, added new sections for Quickstart, Models Supported, What's New, and Start Building. Removed previous sections about Bedrock capabilities and getting started.
Security assessment
No security vulnerabilities or weaknesses are addressed. Changes focus on API usage examples, model listings, and feature updates without security context. The mention of 'secure, enterprise-grade access' is marketing language without technical security documentation.
Diff
diff --git a/bedrock/latest/userguide/what-is-bedrock.md b/bedrock/latest/userguide/what-is-bedrock.md index a328b3178..1460fd405 100644 --- a//bedrock/latest/userguide/what-is-bedrock.md +++ b//bedrock/latest/userguide/what-is-bedrock.md @@ -5 +5 @@ -What can I do with Amazon Bedrock?How do I get started with Amazon Bedrock? +QuickstartModels supportedWhat's new?Start Building @@ -7 +7 @@ What can I do with Amazon Bedrock?How do I get started with Amazon Bedrock? -# What is Amazon Bedrock? +# Overview @@ -9 +9 @@ What can I do with Amazon Bedrock?How do I get started with Amazon Bedrock? -Amazon Bedrock is a fully managed service that makes high-performing foundation models (FMs) from leading AI companies and Amazon available for your use through a unified API. You can choose from a wide range of foundation models to find the model that is best suited for your use case. Amazon Bedrock also offers a broad set of capabilities to build generative AI applications with security, privacy, and responsible AI. Using Amazon Bedrock, you can easily experiment with and evaluate top foundation models for your use cases, privately customize them with your data using techniques such as fine-tuning and Retrieval Augmented Generation (RAG), and build agents that execute tasks using your enterprise systems and data sources. +Amazon Bedrock is a fully managed service that provides secure, enterprise-grade access to [high-performing foundation models](./models.html) from leading AI companies, enabling you to build and scale generative AI applications. @@ -11 +11 @@ Amazon Bedrock is a fully managed service that makes high-performing foundation -With Amazon Bedrock's serverless experience, you can get started quickly, privately customize foundation models with your own data, and easily and securely integrate and deploy them into your applications using AWS tools without having to manage any infrastructure. +## Quickstart @@ -13 +13 @@ With Amazon Bedrock's serverless experience, you can get started quickly, privat -###### Topics +Responses API @@ -15 +14,0 @@ With Amazon Bedrock's serverless experience, you can get started quickly, privat - * What can I do with Amazon Bedrock? @@ -17 +15,0 @@ With Amazon Bedrock's serverless experience, you can get started quickly, privat - * How do I get started with Amazon Bedrock? @@ -19 +17 @@ With Amazon Bedrock's serverless experience, you can get started quickly, privat - * [Amazon Bedrock pricing](./bedrock-pricing.html) + from openai import OpenAI @@ -21 +19 @@ With Amazon Bedrock's serverless experience, you can get started quickly, privat - * [Key terminology](./key-definitions.html) + client = OpenAI() @@ -22,0 +21,5 @@ With Amazon Bedrock's serverless experience, you can get started quickly, privat + response = client.responses.create( + model="openai.gpt-oss-120b", + input="Can you explain the features of Amazon Bedrock?" + ) + print(response) @@ -23,0 +27 @@ With Amazon Bedrock's serverless experience, you can get started quickly, privat +Chat Completions API @@ -26 +29,0 @@ With Amazon Bedrock's serverless experience, you can get started quickly, privat -## What can I do with Amazon Bedrock? @@ -28 +31 @@ With Amazon Bedrock's serverless experience, you can get started quickly, privat -You can use Amazon Bedrock to do the following: + from openai import OpenAI @@ -30 +33 @@ You can use Amazon Bedrock to do the following: - * **Experiment with prompts and configurations** – [Submit prompts and generate responses with model inference](./inference.html) by sending prompts using different configurations and foundation models to generate responses. You can use the API or the text, image, and chat playgrounds in the console to experiment in a graphical interface. When you're ready, set up your application to make requests to the `InvokeModel` APIs. + client = OpenAI() @@ -32 +35,5 @@ You can use Amazon Bedrock to do the following: - * **Augment response generation with information from your data sources** – [Create knowledge bases](./knowledge-base.html) by uploading data sources to be queried in order to augment a foundation model's generation of responses. + response = client.chat.completions.create( + model="openai.gpt-oss-120b", + messages=[{"role": "user", "content": "Can you explain the features of Amazon Bedrock?"}] + ) + print(response) @@ -34 +41 @@ You can use Amazon Bedrock to do the following: - * **Create applications that reason through how to help a customer** – [Build agents](./agents.html) that use foundation models, make API calls, and (optionally) query knowledge bases in order to reason through and carry out tasks for your customers. +Invoke API @@ -36 +42,0 @@ You can use Amazon Bedrock to do the following: - * **Adapt models to specific tasks and domains with training data** – [Customize an Amazon Bedrock foundation model](./custom-models.html) by providing training data for fine-tuning or continued-pretraining in order to adjust a model's parameters and improve its performance on specific tasks or in certain domains. @@ -38 +43,0 @@ You can use Amazon Bedrock to do the following: - * **Improve your FM-based application's efficiency and output** – [Purchase Provisioned Throughput](./prov-throughput.html) for a foundation model in order to run inference on models more efficiently and at discounted rates. @@ -40 +45,2 @@ You can use Amazon Bedrock to do the following: - * **Determine the best model for your use case** – [Evaluate outputs of different models](./evaluation.html) with built-in or custom prompt datasets to determine the model that is best suited for your application. + import json + import boto3 @@ -42 +48,10 @@ You can use Amazon Bedrock to do the following: - * **Prevent inappropriate or unwanted content** – [Use guardrails](./guardrails.html) to implement safeguards for your generative AI applications. + client = boto3.client('bedrock-runtime', region_name='us-east-1') + response = client.invoke_model( + modelId='anthropic.claude-opus-4-6-v1', + body=json.dumps({ + 'anthropic_version': 'bedrock-2023-05-31', + 'messages': [{ 'role': 'user', 'content': 'Can you explain the features of Amazon Bedrock?'}], + 'max_tokens': 1024 + }) + ) + print(json.loads(response['body'].read())) @@ -44 +59 @@ You can use Amazon Bedrock to do the following: - * **Optimize your FM's latency** – [Get faster response times and improved responsiveness](./latency-optimized-inference.html) for AI applications with Latency-optimized inference for foundation models. +Converse API @@ -46 +60,0 @@ You can use Amazon Bedrock to do the following: -###### Note @@ -48 +61,0 @@ You can use Amazon Bedrock to do the following: -The Latency Optimized Inference feature is in preview release for Amazon Bedrock and is subject to change. @@ -49,0 +63 @@ The Latency Optimized Inference feature is in preview release for Amazon Bedrock + import boto3 @@ -50,0 +65,11 @@ The Latency Optimized Inference feature is in preview release for Amazon Bedrock + client = boto3.client('bedrock-runtime', region_name='us-east-1') + response = client.converse( + modelId='anthropic.claude-opus-4-6-v1', + messages=[ + { + 'role': 'user', + 'content': [{'text': 'Can you explain the features of Amazon Bedrock?'}] + } + ] + ) + print(response) @@ -51,0 +77 @@ The Latency Optimized Inference feature is in preview release for Amazon Bedrock +Read the [Quickstart](./getting-started.html) to write your first API call using Amazon Bedrock in under five minutes. @@ -53 +79 @@ The Latency Optimized Inference feature is in preview release for Amazon Bedrock -To learn about Regions that support Amazon Bedrock and the foundation models and features that Amazon Bedrock supports, see [Supported foundation models in Amazon Bedrock](./models-supported.html) and [Feature support by AWS Region in Amazon Bedrock](./features-regions.html). +## Models supported @@ -55 +81 @@ To learn about Regions that support Amazon Bedrock and the foundation models and -## How do I get started with Amazon Bedrock? +Bedrock supports [100+ foundation models](./models.html) from industry-leading providers, including Amazon, Anthropic, DeepSeek, Moonshot AI, MiniMax, and OpenAI. @@ -57 +83,2 @@ To learn about Regions that support Amazon Bedrock and the foundation models and -We recommend that you start with Amazon Bedrock by doing the following: + **Nova 2 Pro** |  **Claude Opus 4.6** |  **Deepseek 3.2** |  **Kimi K2.5** |  **MiniMax M2.1** |  **GPT-OSS-20B** +---|---|---|---|---|--- @@ -59 +86 @@ We recommend that you start with Amazon Bedrock by doing the following: - 1. Familiarize yourself with the [terms and concepts](./key-definitions.html) that Amazon Bedrock uses. +## What's new? @@ -61 +88 @@ We recommend that you start with Amazon Bedrock by doing the following: - 2. Understand how AWS [charges](./bedrock-pricing.html) you for using Amazon Bedrock. + * **[Six new open weight models](https://aws.amazon.com/about-aws/whats-new/2026/02/amazon-bedrock-adds-support-six-open-weights-models/)** : Amazon Bedrock now supports six new models spanning frontier reasoning and agentic coding: DeepSeek V3.2, MiniMax M2.1, GLM 4.7, GLM 4.7 Flash, Kimi K2.5, and Qwen3 Coder Next. @@ -63 +90 @@ We recommend that you start with Amazon Bedrock by doing the following: - 3. Try the [Get started with Amazon Bedrock](./getting-started.html) tutorials. In the tutorials, you learn how to use the playgrounds in [Amazon Bedrock console](). You also learn and how to use the [AWS SDK](./getting-started-api.html) to call Amazon Bedrock API operations. + * **Claude 4.6[ now available](https://aws.amazon.com/about-aws/whats-new/2026/2/claude-opus-4.6-available-amazon-bedrock/)**: According to Anthropic, Opus 4.6 is their most intelligent model and the world's best model for coding, enterprise agents, and professional work. Read more here. @@ -65 +92 @@ We recommend that you start with Amazon Bedrock by doing the following: - 4. Read the documentation for the features that you want to include in your application. + * **Server-side tools** : Amazon Bedrock [now supports](https://aws.amazon.com/about-aws/whats-new/2026/01/amazon-bedrock-server-side-custom-tools-responses-api/) server-side tools in the Responses API using OpenAI API-compatible service endpoints. @@ -66,0 +94 @@ We recommend that you start with Amazon Bedrock by doing the following: + * **1-hour prompt caching duration** : Amazon Bedrock [now supports](https://aws.amazon.com/about-aws/whats-new/2026/01/amazon-bedrock-one-hour-duration-prompt-caching/) a 1-hour time-to-live (TTL) option for prompt caching for select Anthropic Claude models. @@ -67,0 +96 @@ We recommend that you start with Amazon Bedrock by doing the following: + * **NVIDIA Nemotron 3 Nano[ now available](https://aws.amazon.com/about-aws/whats-new/2025/12/nvidia-nemotron-3-nano-amazon-bedrock/)**: NVIDIA Nemotron 3 Nano 30B A3B delivers high reasoning performance, native tool calling support, and extended context processing with 256k token context window. @@ -69,0 +99,9 @@ We recommend that you start with Amazon Bedrock by doing the following: + + +## Start Building + + | Explore the [APIs supported by Amazon Bedrock](./apis.html) and [Endpoints supported by Amazon Bedrock](./endpoints.html) supported by Amazon Bedrock. +---|--- + | Build using the [Submit prompts and generate responses with model inference](./inference.html) operations provided by Amazon Bedrock. + | Customize your models to improve performance and quality. [Customize your model to improve its performance for your use case](./custom-models.html) + @@ -76 +114 @@ To use the Amazon Web Services Documentation, Javascript must be enabled. Please -Amazon Bedrock pricing +Quickstart