How to use DALLE using Azure Open AI?

[blacksmithop]

I found the docs for dalle_image_generator but could not find anything for Azure OpenAI

[dosubot[bot]]

To use DALL-E with Azure OpenAI in LangChain, you can follow these steps:

1. Set Environment Variables:

   • Configure your environment with the necessary API details for Azure OpenAI.

   export OPENAI_API_VERSION=2023-12-01-preview
   export AZURE_OPENAI_ENDPOINT=https://your-resource-name.openai.azure.com
   export AZURE_OPENAI_API_KEY=<your Azure OpenAI API key>

2. Authenticate (Optional):

   • If you prefer using Azure Active Directory (AAD) for authentication, use the azure-identity package.

   import os
   from azure.identity import DefaultAzureCredential
   
   credential = DefaultAzureCredential()
   os.environ["OPENAI_API_TYPE"] = "azure_ad"
   os.environ["OPENAI_API_KEY"] = credential.get_token("https://cognitiveservices.azure.com/.default").token

3. Use the DALL-E API Wrapper:

   • LangChain provides a DallEAPIWrapper class for integrating with DALL-E. You can use this class to generate images by providing a text prompt.

   from langchain_community.utilities.dalle_image_generator import DallEAPIWrapper
   
   dalle = DallEAPIWrapper()
   image = dalle.generate_image(prompt="A futuristic cityscape")

These steps will help you configure and use DALL-E with Azure OpenAI in LangChain. If you need more detailed guidance, you might want to check the LangChain documentation or the specific module for DALL-E integration ^[1][2].

To continue talking to Dosu, mention @dosu.

---

^{Help Dosu learn! Give it feedback: Great Response | Irrelevant Answer | Incorrect Sources | Too Verbose | Hallucination | Other}

[blacksmithop]

Here's what worked for me (non-langchain solution)

from openai import AzureOpenAI
import json

client = AzureOpenAI(
    api_version = OPENAI_API_VERSION,
    api_key = AZURE_OPENAI_API_KEY,
    azure_endpoint = AZURE_OPENAI_ENDPOINT,
)

response = client.images.generate(
     model = "dall-e-3",
     prompt = prompt,
)

json_response = json.loads(response.model_dump_json())
image_url = json_response["data"][0]["url"]