Question 241910 in Azure Question posted in
The official documentation to get you started can be found here.

PermissionError while executing langchain with AzureOpenAI

I’m working with AzureOpenAI and langchain, constantly getting hit by PermissionError. This mostly could be due to the proxy, but can someone please check the code —

from langchain.llms import OpenAI, AzureOpenAI
from langchain.prompts import PromptTemplate
from langchain.chains import LLMChain

llm = AzureOpenAI(openai_api_type="", openai_api_base="", deployment_name="", model_name="", openai_api_key="", openai_api_version="")

template = """"
Translate the following text from {source_lang} to {dest_lang}: {source_text}
"""

prompt_name = PromptTemplate(input_variables=["source_lang", "dest_lang", "source_text"], template=template)
chain = LLMChain(llm=llm, prompt=prompt_name)

chain.predict(source_lang="English", dest_lang="Spanish", source_text="How are you?")

chain(inputs={"source_lang": "English", "dest_lang": "Spanish", "source_text": "How are you"})

I also tried the additional openai_proxy parameter without much luck.

Tags:

2

Answers


  1. 0

    enter image description here

    your code runs smoothly! Below my llm config. Other variables are set with .env

    llm=AzureChatOpenAI(
    deployment_name=AZURE_OPENAI_CHATGPT_DEPLOYMENT,
    temperature=0.0,
)
    Login or Signup to reply.
  2. 0

    To fix this permission error, you need to create a service principal and assign it the appropriate permissions. Follow the below steps to create a service principal,

    • Go to the Azure portal and create a new service principal.

    • In the service principal blade, select the "API permissions" tab.

    • Click on the "Add permission" button and select the "Azure OpenAI" API.

    • Select the "Delegated permissions" tab and select the following permissions:

      openai.runtime.invoke
openai.runtime.models.list

    • Click on the "Save" button.

    • Set the following environment variables in your operating system’s environment variables or in a .env file.

      AZURE_OPENAI_CLIENT_ID=your-client-id
AZURE_OPENAI_CLIENT_SECRET=your-client-secret
AZURE_OPENAI_TENANT_ID=your-tenant-id
    Login or Signup to reply.
Please signup or login to give your own answer.

Back To Top
Search