AWS solutions documentation change
Summary
Updated documentation with typo fixes, expanded model provider options (Bedrock/SageMaker), clarified RAG configurations, and improved prompt configuration details. Added descriptions of Bedrock model types (Quick Start Models, Other Foundation Models, Inference Profiles, Provisioned Models).
Security assessment
Changes primarily involve typo corrections, feature clarifications, and expanded configuration options without addressing specific security vulnerabilities. While the documentation mentions security-related components like VPCs and security groups, these are standard network configuration elements rather than new security advisories or vulnerability fixes. The RBAC mention in Kendra configurations was already present and only received a typo fix.
Diff
diff --git a/solutions/latest/generative-ai-application-builder-on-aws/step-3-deploy-a-use-case-using-deployment-dashboard-wizard.md b/solutions/latest/generative-ai-application-builder-on-aws/step-3-deploy-a-use-case-using-deployment-dashboard-wizard.md index e8b1dacb1..d62c5aead 100644 --- a//solutions/latest/generative-ai-application-builder-on-aws/step-3-deploy-a-use-case-using-deployment-dashboard-wizard.md +++ b//solutions/latest/generative-ai-application-builder-on-aws/step-3-deploy-a-use-case-using-deployment-dashboard-wizard.md @@ -68 +68 @@ Additionally, if you need to add more users to a use case, refer to the [Managin -### Select network configuration +### Select network configuration @@ -70 +70 @@ Additionally, if you need to add more users to a use case, refer to the [Managin -This wizard step allows you to deploy the use case with a pre-existing or new [Amazon Virtual Private Cloud](https://aws.amazon.com/vpc/) (Amazon VPC). If selecting pre-existing VPC, you are required to provide a VPC ID, up to 16 subnet Ids and up to 5 security group IDs to use with this VPC. If you’re not using a pre-existing VPC, these settings will be configured for you. +This wizard step allows you to deploy the use case with a pre-existing or new [Amazon Virtual Private Cloud](https://aws.amazon.com/vpc/) (Amazon VPC). If selecting pre-existing VPC, you are required to provide a VPC ID, up to 16 subnet Ids and up to 5 security group IDs to use with this VPC. If you’re not using a pre-existing VPC, these settings will be configured for you. @@ -74 +74,16 @@ This wizard step allows you to deploy the use case with a pre-existing or new [A -In the **Select model** step, you can choose your model provider, such as **Amazon Bedrock** , and select a model from the available models names. Alternatively, you can create a SageMaker AI model endpoint in the SageMaker AI console and provide the input schema that the model expects and output JSONPath for the LLM response. You can refer to the [Using Amazon SageMaker as an LLM Provider](./use-the-solution.html#using-amazon-sagemaker-as-an-llm-provider) section and [SageMaker payload examples](https://github.com/aws-solutions/generative-ai-application-builder-on-aws/tree/main/docs/sagemaker-payload-examples) provided in the solution’s GitHub repository. +In the **Select model** step, you can choose your model provider from the dropdown menu. There are two options: **Bedrock** and **SageMaker**. + +If you select **SageMaker** , you can create a SageMaker AI model endpoint in the SageMaker AI console and provide the input schema that the model expects and output JSONPath for the LLM response. You can refer to the [Using Amazon SageMaker AI as an LLM Provider](./use-the-solution.html#using-amazon-sagemaker-ai-as-an-llm-provider) section and [SageMaker AI payload examples](https://github.com/aws-solutions/generative-ai-application-builder-on-aws/tree/main/docs/sagemaker-payload-examples) provided in the solution’s GitHub repository. + +If you select **Amazon Bedrock** , you will be presented with four options: + + * **Quick Start Models** \- Get started quickly with a collection of models with different price/performance characteristics. Recommended for building your first apps. This option allows you to select a model name from the provided list. + + * **Other Foundation Models** \- Access the full range of foundation models with different capabilities and specializations. This option allows you to enter the model ID for your desired Bedrock on-demand foundation model. + + * **Inference Profiles** \- Inference profiles leverage Bedrock’s cross-region inference to increase throughput and improve resiliency by routing your requests across multiple AWS Regions during peak utilization bursts. This option allows you to enter the ID of the inference profile you want to use. + + * **Provisioned Models** \- Dedicated throughput capacity for production workloads requiring consistent performance. This option allows you to enter the ARN of the provisioned/custom model to use from Amazon Bedrock. + + + @@ -94 +109 @@ After you identify the inference profile ID to use, you can use this during the - 2. Select **cross-region inference** (instead of an on-demand model). + 2. Select the **Inference Profiles** radio button option. @@ -96 +111 @@ After you identify the inference profile ID to use, you can use this during the - 3. Enter your inference profile ID in the text box. + 3. Enter your inference profile ID in the text box that appears. @@ -107 +122 @@ If you’re looking to deploy a non-Retrieval Augmented Generation (RAG) use cas -However, if you wish to enable RAG as a part of your deployment, you can now provide either a pre-configured _Amazon Kendra Index Id_ or an _Amazon Bedrock Knowledge Base ID_. You can also create a new Amazon Kendra Index for use with the solution. The solution currently supports Amazon Kendra and Amazon Bedrock Knowledge Bases as knowledge bases for your RAG-based use case deployment. +However, if you wish to enable RAG as a part of your deployment, you can now provide either a pre-configured _Amazon Kendra Index Id_ or an _Amazon Bedrock Knowledge Base ID_. You can also create a new Amazon Kendra Index for use with the solution. The solution currently supports Amazon Kendra and Amazon Bedrock Knowledge Bases as knowledge bases for your RAG-based use case deployment. @@ -109 +124 @@ However, if you wish to enable RAG as a part of your deployment, you can now pro -Refer to the [Configuring a Knowledge Base](./configuring-a-knowledge-base.html) section for guidelines on ingesting data into the knowledge base for use with your RAG-based deployment. +Refer to the [Configuring a Knowledge Base](./configuring-a-knowledge-base.html) section for guidelines on ingesting data into the knowledge base for use with your RAG-based deployment. @@ -111 +126 @@ Refer to the [Configuring a Knowledge Base](./configuring-a-knowledge-base.html -#### Advanced RAG configurations +#### Advanced RAG configurations @@ -113 +128 @@ Refer to the [Configuring a Knowledge Base](./configuring-a-knowledge-base.html -The wizard allows you to select advanced options for use with your RAG deployment such as the **number of documents to retrieve** each time a query is sent to your knowledge base, a **static text response** from the LLM when no documents are found in the knowledge base, whether you wish to **display document sources** with your LLM response for sanity checks, etc. You can additionally also configure knowledge base specific configurations for Amazon Kendra such as [Role-based Access Control (RBAC)](https://docs.aws.amazon.com/kendra/latest/dg/create-index-access-control.html), or [Override Search Type](https://docs.aws.amazon.com/bedrock/latest/userguide/kb-test-config.html) when using Amazon OpenSearch Serverless with Amazon Bedrock Knowledge Bases. Refer to the [Advanced Knowledge Base settings](./advanced-knowledge-base-settings.html) section for more details on these advanced settings. +The wizard allows you to select advanced options for use with your RAG deployment such as the **number of documents to retrieve** each time a query is sent to your knowledge base, a **static text response** from the LLM when no documents are found in the knowledge base, whether you wish to **display document sources** with your LLM response for sanity checks, etc. You can additionally also configure knowledge base specific configurations for Amazon Kendra such as [Role-based Access Control (RBAC)](https://docs.aws.amazon.com/kendra/latest/dg/create-index-access-control.html), or [Override Search Type](https://docs.aws.amazon.com/bedrock/latest/userguide/kb-test-config.html) when using Amazon OpenSearch Serverless with Amazon Bedrock Knowledge Bases. Refer to the [Advanced Knowledge Base settings](./advanced-knowledge-base-settings.html) section for more details on these advanced settings. @@ -121 +136,10 @@ Your knowledge base must be in the same account and Region as the deployed Deplo -In this step, you can configure your prompt for use with the LLM. Each prompt requires a minimum of two placeholders - `{input}` and `{history}`. For RAG use cases, the `{context}` placeholder is additionally required. These placeholders instruct the LLM on where to draw user input, conversation history, and information retrieved from the knowledge base from. +In this step, you can configure your prompt for use with the LLM. Prompts may require placeholders such as `{input}`, `{history}` and `{context}`. These placeholders instruct the LLM on where to draw user input, conversation history, and information retrieved from the knowledge base from. + + * For Bedrock model provider, the system prompt must be provided which has no restrictions for a non-RAG use-case. The disambiguation prompt for Bedrock model provider however, requires a minimum of two placeholders - `{input}` and `{history}` + + * For SageMaker model provider, system and disambiguation prompts, both require a minimum of two placeholders - `{input}` and `{history}`. + + * For RAG use cases, for each model provider, the `{context}` placeholder is additionally required. + + + @@ -123 +147 @@ In this step, you can configure your prompt for use with the LLM. Each prompt r -For more information, see [Configuring your prompts](./configuring-your-prompts.html). You can also refer to the [Tips for managing model token limits](./tips-for-managing-model-token-limits.html) section while selecting token limit sizes for your prompts. +For more information, see [Configuring your prompts](./configuring-your-prompts.html). You can also refer to the [Tips for managing model token limits](./tips-for-managing-model-token-limits.html) section while selecting token limit sizes for your prompts. @@ -163 +187 @@ This step is the same as the Text use case described previously. -#### Select network configuration +#### Select network configuration