Generative Artificial Intelligence (AI) on AWS

Learn, Build, and Explore with Amazon Bedrock, CodeWhisperer, and more

Generative Artificial Intelligence (AI) on AWS

Learn, Build, and Explore with Amazon CodeWhisperer, Amazon Bedrock, and more...

Generative AI - A Primer

Generative AI is everywhere now-a-days. But what is it? Well, think of it as a type of AI that can create new content an ideas, including conversations, stories, images, videos, and music. For your gen AI journey with AWS, this video gives a foundational overview of where it came from, where it’s going, how it works, and how to get started.

Build your Generative AI Applications on AWS

To build generative AI applications on AWS, you can start with Amazon Bedrock to choose the right Foundation Model (FM) for your use-case. If you prefer, you can also use Amazon SageMaker JumpStart’s ML Hub to accelerate your model development. The selected model can then be customized with additional training in AWS to suit the application’s needs. And as you program and code, the Amazon CodeWhisperer service can help as your developer coding tool.

PartyRock, an Amazon Bedrock Playground

PartyRock, powered by Amazon Bedrock, is an engaging and user-friendly generative AI app-building playground. Within seconds, you can craft your unique apps, share them, and delve into the world of generative AI—all while enjoying the experience.

Amazon Bedrock and Foundation Models (FMs)

Amazon Bedrock is a fully managed service that lets you choose the FM that's best suited for your use-case. There are a broad set of FM's available including Amazon's Titan, and from leading Al startups - Al21Labs, Anthropic, co:here, Meta, Stability.Al

Amazon SageMaker JumpStart ML Hub

Amazon SageMaker JumpStart has hundreds of built-in algorithms and pre-trained models to accelerate ML model building and deployment within SageMaker, while ensuring data security. The broad set of FMs from Amazon, proprietary, and publicly available, represent top-scoring models via Holistic Evaluation of Language Models (HELM) benchmarks.

Amazon CodeWhisperer, your Al coding companion

Amazon CodeWhisperer generates code suggestions in real-time based on natural language comments, and prior code in the Integrated Development Environment (IDE). It works with many different programming languages such as Python, Java, and JavaScript, and IDEs including VS Code, IntelliJ IDEA, and from AWS such as JupyterLab.

Code with Amazon CodeWhisperer

Being trained on AWS data and APIs, Amazon CodeWhisperer analyzes existing code in the IDE (whether generated by CodeWhisperer or written by you), identifies problematic code with high accuracy, and provides intelligent suggestions on how to remediate it. With the customization capability more precise suggestions are generated, by including your organization’s internal APIs, libraries, classes, methods, and best practices. Explore with Amazon CodeWhisperer and get a productivity boost.

Build on Amazon Bedrock

With Amazon Bedrock’s comprehensive capabilities, you can experiment with a variety of Foundation Models (FMs), customize them using your data with techniques such as Fine-Tuning and Retrieval-Augmented Generation (RAG), and create managed agents. Explore now with Amazon Bedrock to execute your complex business tasks—from booking travel and processing insurance claims to creating ad campaigns and managing inventory—all without writing any code. 

Click image to enlarge

Amazon Titan

Amazon Titan are a family of pretrained foundation models by AWS on large datasets, making them powerful, general-purpose models built to support a variety of use cases. Use them as is or privately customize them with your own data

Developer Experience

Amazon Bedrock makes it easy for developers to work with a broad range of high-performing foundation models (FMs)

Amazon Bedrock API

Get more detailed information about the Bedrock API actions and their parameters

Integrating FMs into your code with Amazon Bedrock

Discover Amazon Bedrock and learn how to integrate gen AI models from leading Al startups and Amazon into your application

Train generative AI Models on Purpose-built Accelerators

Whatever customers are trying to do with FMs—running them, building them, customizing them—they need the most performant, cost-effective, purpose-built ML infrastructure. Over the past decade, AWS has been investing with our partners and silicon to offer a broad choice of high-performance, low-cost ML infrastructure chip options. The AWS Trainium and AWS Inferentia chips offer the lowest cost for generative AI training models and running inference in the cloud

AWS Trainium

Up to 50% savings on training costs over comparable Amazon EC2 instances

AWS Inferentia2

Up to 40% better price performance than comparable Amazon EC2 instances

AWS ML Infrastructure

Accelerate your ML innovation, AWS offers the ideal combination of high performance, cost-effective, and energy-efficient purpose-built ML tools and accelerators, optimized for ML applications.

Try Some Sample Apps

To build generative AI applications on AWS, you start with Amazon CodeWhisperer as your developer coding tool, and then use Amazon Bedrock to choose the right Foundation Model (FM) for your user-case. If you prefer, you can also use Amazon SageMaker JumpStart’s ML Hub to accelerate your model development. The selected model can then be customized with additional training in AWS to suit the application’s needs.


Build a GenAl app quickly with FMs

With Streamlit (open-source Python library), Python, Claude, Stable Diffusion, and Amazon Bedrock. Explore four distinct use cases, from image generation to text summarization, demonstrating the versatility of this new service.


GenAl in the Retail Industry

From virtual try-on experiences to crafting personalized advertisements, the impact of Stable Diffusion is far-reaching.


Maintain a Coherent Conversation with LLMs

Use LangChain with Amazon Bedrock and Amazon DynamoDB and to build applications to engage in Natural dialogue


Build Your Own Knowledge Base with Multilingual Q&A

Use Amazon Kendra, Amazon Translate, Amazon Comprehend and Amazon SageMaker JumpStart to build a multilingual knowledge base that can summarize search results.

Build Responsibly with AI

Responsible use of AI and ML is key to tackling some of humanity’s most challenging problems, augmenting human performance, and maximizing productivity. AWS is committed to developing fair and accurate AI and ML services and providing you with the tools and guidance needed to build AI and ML applications responsibly. 

More Resources


Prompt Engineering for Developers

Learn effective prompt engineering techniques for Large Language Models (LLMs), to facilitate software architecture decisions.
Visit Post 

Al-Powered Development: Creating Secure and Resilient Apps

Learn about advances in Al developer tools such as Amazon CodeWhisperer and Amazon CodeGuru, and how they can help increase your applications' security, availability, and resiliency.
Visit Post 

Choosing an AWS ML service

Decision Guide - Pick the right AI and ML services, frameworks, and foundation models to support your work
Visit Post 

Generative AI with Large Language Models Course by DeepLearning.AI and AWS

This new training on Coursera, developed with AWS experts and AI educators like Dr. Andrew Ng, helps you from learning about LLMs to learn about LLMs to deploying them for real world applications
Visit Post 


Discover, Explore, and Apply the collection of developer focused content, resources, and hands-on tools that can make a difference in your cloud computing journey.

Was this page helpful?