Does dify now support azure openai gpt 4o?

[lx1054331851]

Self Checks

• This is only for bug report, if you would like to ask a question, please head to Discussions.
• I have searched for existing issues search for existing issues, including closed ones.
• I confirm that I am using English to submit this report (我已阅读并同意 Language Policy).
• Please do not modify this template :) and fill in all the required fields.

Dify version

0.6.8

Cloud or Self Hosted

Self Hosted (Docker)

Steps to reproduce

I can not add azure openai gpt 4o model.

✔️ Expected Behavior

No response

❌ Actual Behavior

No response

[dosubot]

Yes, Dify version 0.6.8, self-hosted with Docker, supports the Azure OpenAI GPT-4o model.

To add the Azure OpenAI GPT-4o model, follow these steps:

1. Obtain the API Key: Get the API key from the Azure OpenAI Service.

2. Configure Dify:

   • Go to Settings > Model Provider in the Dify interface.
   • Select Azure OpenAI Service and enter the API key.

This will integrate the Azure OpenAI GPT-4o model into your Dify setup. The model supports various features such as agent thought, vision, multi-tool call, and stream tool call, and allows for customizable parameters like temperature, top_p, presence_penalty, frequency_penalty, and max_tokens.

Additionally, you can validate the credentials and invoke the model using the following Python code snippet:

import os
from core.model_runtime.model_providers.azure_openai.llm.llm import AzureOpenAILargeLanguageModel

model = AzureOpenAILargeLanguageModel()

# Validate credentials
model.validate_credentials(
    model='gpt-4o',
    credentials={
        'openai_api_base': os.environ.get('AZURE_OPENAI_API_BASE'),
        'openai_api_key': os.environ.get('AZURE_OPENAI_API_KEY'),
        'base_model_name': 'gpt-4o'
    }
)

# Invoke the model
result = model.invoke(
    model='gpt-4o',
    credentials={
        'openai_api_base': os.environ.get('AZURE_OPENAI_API_BASE'),
        'openai_api_key': os.environ.get('AZURE_OPENAI_API_KEY'),
        'base_model_name': 'gpt-4o'
    },
    prompt_messages=[
        UserPromptMessage(
            content='Hello World!'
        )
    ],
    model_parameters={
        'temperature': 0.0,
        'max_tokens': 100
    },
    stream=False,
    user="abc-123"
)

assert isinstance(result, LLMResult)
assert len(result.message.content) > 0

This code snippet demonstrates how to validate the credentials and invoke the Azure OpenAI GPT-4o model within Dify. For more detailed information, you can refer to the Dify documentation and the relevant Dify repository files.

Details

Sources

• Added Azure OpenAI Service Model, error when selecting base model "gpt-4-turbo-2024-04-09"
• Add support for Azure openai 1106 model
• gpt4 doesn't exist
• Azure OpenAI支持問題
• en/tutorials/model-configuration/README.md
• api/core/model_runtime/model_providers/azure_openai/_constant.py
• api/tests/integration_tests/model_runtime/azure_openai/test_llm.py

About Dosu
This response is meant to be useful and save you time. It is not meant to be a precise solution, but rather a starting point for your own research.

Help me be more useful! Please leave a 👍 if this is helpful and 👎 if it is irrelevant.

To continue the conversation, mention @dosu.

[brightwang-igg]

d5a33a0
Pull the new code from the main branch to support Azure GPT-4o."