Create an OpenAI resource and a deployment in Azure Portal - Instructions here
For the Type, choose option 1: Allow all networks
Before you begin make sure to set the following environment variables and values in application.properties
Add the OpenAI API Key and Endpoint from the Azure Portal > < Your OpenAI Resource> > Keys and Endpoint
export SPRING_AI_AZURE_OPENAI_API_KEY=<INSERT KEY HERE>
export SPRING_AI_AZURE_OPENAI_ENDPOINT=<INSERT ENDPOINT URL HERE>Next, set the following environment variables and values in application.properties to match your deployment in Open AI Studio. In this example, the Deployment name is gpt-35-turbo-16k and the Model name is gpt-35-turbo-16k.
spring.ai.azure.openai.model=gpt-35-turbo-16k
spring.ai.azure.openai.temperature=0.7
spring.ai.azure.openai.deploymentOrModelId=gpt-35-turbo-16kThe workshop consists of six examples, each with a dedicated README file.
All six workshop examples are organized into individual Java packages within this project. In each package, you'll find a Spring @RestController class that serves as the entry point for showcasing the discussed functionality.
To interact with the @RestController, you will be using the http utility as a user-friendly alternative to curl.
Detailed instructions and exercises for each example can be found in their respective README files:
- 1-README-tell-me-a-joke.md
- 2-README-prompt-templating.md
- 3-README-prompt-roles.md
- 4-README-output-parser.md
- 5-README-stuff-prompt.md
- 6-README-retrieval-augmented-generation.md
These guides will walk you through the workshop exercises.