Amazon Bedrock is a fully managed service that offers API access to a choice of high-performing foundation models (FMs) from leading AI companies including AI21 Labs, Anthropic, Cohere, Meta, Stability AI, and Amazon, along with a broad set of capabilities that you need to build generative AI applications, simplifying development while maintaining privacy and security.
This demo provides a basic introduction to some GenAI use cases, by allowing you to interact with FMs using Amazon Bedrock. This demo application is intended for quick deployment on your workstation to allow you explore Amazon Bedrock easily.
Watch a video of the demo below (the Chrome browser is recommended)!
amazon_bedrock_intro_demo_video.mp4
NOTE: Refer to the Amazon Bedrock pricing to understand the costs incurred with using Amazon Bedrock.
- AWS account (sandbox account recommended)
- IAM user or role with Administrator access or the required permissions to access Amazon Bedrock and its FMs. Configure this principal's credentials in your environment's default AWS profile (AWS_PROFILE). Also, ensure that you have enabled Model access on Amazon Bedrock.
- Python 3.9+
- Linux (preferred)
- Internet access
Implement the following steps to start using the demo application:
STEP 1: Create a Python virtual environment. This is optional depending on your environment.
python -m venv demo
cd demo
source bin/activate
STEP 2: Clone the git repository
git clone https://github.com/aws-samples/amazon-bedrock-intro-demo.git
Alternatively, download the code and extract the amazon-bedrock-intro-demo directory.
STEP 3: Install the required python modules
cd amazon-bedrock-intro-demo
pip install -r requirements.txt
STEP 4: Launch the Streamlit application and access the displayed URL
streamlit run main.py
NOTE: This demo is not maintained. For any updates, go here
