Create a file .env and write the following env variables:
AZURE_OPENAI_ENDPOINT=https://<REPLACE_ME>/api/v1
AZURE_OPENAI_API_KEY=<REPLACE_ME>
OPENAI_API_VERSION=<REPLACE_ME>
Note: The way to access those credentials will be sent through private communication.
This repository contains the following samples
| Sample | Explanation |
|---|---|
| dalle3.py | Generate an image using Azure OpenAI DALL-E 3 |
| gpt.py | Generate text using a Azure OpenAI GPT model |
| embeddings.py | Generate embeddings using an Azure OpenAI embedding model |
| truncate_tokens.py | Shows how to truncate tokens to control context size usage and avoid going over the model's limit |
| functions_calling.py | Shows how to use Function calling |
| Model name | Type | Sample |
|---|---|---|
| Dalle3 | OpenAI Dalle3 | dalle3.py |
| gpt-35-turbo | OpenAI Chat | gpt.py |
| gpt-4-turbo | OpenAI Chat | gpt.py |
| gpt-4-vision | OpenAI Chat | gpt.py |
| text-embedding-ada-002 | OpenAI Embedding | embeddings.py |
Each sample reads Azure OpenAI connection information from environment variables stored in the .env file using the following snippet of code:
from dotenv import load_dotenv
load_dotenv() # take environment variables from .env.