Azure OpenAI

Helicone can be seamlessly integrated with your Azure-OpenAI deployment. This integration requires no additional code changes beyond importing the OpenAI package with Helicone.

MODEL OVERRIDE

When using Azure, the model may sometimes display differently than expected. We have implemented a series of logic to parse out the model, but the returned body of the model can be inconsistent in most cases.

Helicone-Model-Override: [MODEL_NAME]

Click here to learn more about model override

Please ensure to include the ‘api-version’ in all of your requests.

OpenAI v1+

It is mandatory to include the ‘api-key’ in both locations.

  client = OpenAI(
    api_key="[AZURE_OPENAI_API_KEY]",
    base_url="https://oai.helicone.ai/openai/deployments/[DEPLOYMENT]",
    default_headers={
        "Helicone-OpenAI-Api-Base": "https://[AZURE_DOMAIN].openai.azure.com",
        "Helicone-Auth": "Bearer [HELICONE_API_KEY]",
        "api-key": "[AZURE_OPENAI_API_KEY]",
    },
    default_query={
        "api-version": "[API_VERSION]"
    
    }
)

OpenAI <v1

Modify the API base and add a Helicone-Auth header

import openai

openai.api_base = "https://oai.helicone.ai"
openai.api_key = "[AZURE_OPENAI_API_KEY]"
openai.api_version = "[API_VERSION]"
openai.api_type = "azure"

# Create the completion request
openai.ChatCompletion.create(
    model="[DEPLOYMENT]",
    messages=[{"role": "User", "content": "Say hi!"}],
    headers={
        "Helicone-Auth": "Bearer [HELICONE_API_KEY]",
        "Helicone-OpenAI-Api-Base": "https://[AZURE_DOMAIN].openai.azure.com",
    }
)

Congratulations! Your Azure OpenAI requests are now configured to log results to Helicone.

Please ensure to include the ‘api-version’ in all of your requests.

OpenAI v1+

It is mandatory to include the ‘api-key’ in both locations.

  client = OpenAI(
    api_key="[AZURE_OPENAI_API_KEY]",
    base_url="https://oai.helicone.ai/openai/deployments/[DEPLOYMENT]",
    default_headers={
        "Helicone-OpenAI-Api-Base": "https://[AZURE_DOMAIN].openai.azure.com",
        "Helicone-Auth": "Bearer [HELICONE_API_KEY]",
        "api-key": "[AZURE_OPENAI_API_KEY]",
    },
    default_query={
        "api-version": "[API_VERSION]"
    
    }
)

OpenAI <v1

Modify the API base and add a Helicone-Auth header

import openai

openai.api_base = "https://oai.helicone.ai"
openai.api_key = "[AZURE_OPENAI_API_KEY]"
openai.api_version = "[API_VERSION]"
openai.api_type = "azure"

# Create the completion request
openai.ChatCompletion.create(
    model="[DEPLOYMENT]",
    messages=[{"role": "User", "content": "Say hi!"}],
    headers={
        "Helicone-Auth": "Bearer [HELICONE_API_KEY]",
        "Helicone-OpenAI-Api-Base": "https://[AZURE_DOMAIN].openai.azure.com",
    }
)

Congratulations! Your Azure OpenAI requests are now configured to log results to Helicone.

Please ensure to include the ‘api-version’ in all of your requests.

OpenAI v4+

const openai = new OpenAI({
  baseURL: "https://oai.helicone.ai/openai/deployments/[DEPLOYMENTNAME]",
  defaultHeaders: {
    "Helicone-Auth": `Bearer [HELICONE_API_KEY]`,
    "Helicone-OpenAI-API-Base": "https://[AZURE_DOMAIN].openai.azure.com",
    "api-key": "[AZURE_API_KEY]",
  },
  defaultQuery: { "api-version": "[API_VERSION]" },
});

OpenAI <v4

import { Configuration, OpenAIApi } from "openai";

const configuration = new Configuration({
  apiKey: "[AZURE_OPENAI_API_KEY]",
  basePath: "https://oai.helicone.ai/openai/deployments/[DEPLOYMENT]",
  baseOptions: {
    headers: {
      "Helicone-Auth": `Bearer [HELICONE_API_KEY]`,
      "api-key": "[AZURE_OPENAI_API_KEY]",
      "Helicone-OpenAI-Api-Base": "https://[AZURE_DOMAIN].openai.azure.com",
    },
    params: {
      "api-version": "[API_VERSION]",
    },
  },
});

const openai = new OpenAIApi(configuration);

from helicone.openai_proxy import openai
openai.api_type = "azure"
openai.api_base = "https://[YOUR_AZURE_DOMAIN].openai.azure.com"
openai.api_version = "2023-03-15-preview"
openai.api_key = YOUR_AZURE_API_KEY

response = openai.ChatCompletion.create(
engine = 'gpt-35-turbo',
messages = [{
    'role': 'user',
    'content': "Hello World!"
}],
max_tokens=15,
)

const model = new ChatOpenAI({
  azureOpenAIApiKey: "[AZURE_OPENAI_API_KEY]",
  azureOpenAIApiDeploymentName: "openai/deployments/gpt-35-turbo",
  azureOpenAIApiVersion: "2023-03-15-preview",
  azureOpenAIBasePath: "https://oai.helicone.ai",
  configuration: {
    organization: "[organization]",
    baseOptions: {
      headers: {
        "Helicone-Auth": `Bearer ${heliconeApiKey}`,
        "Helicone-OpenAI-Api-Base":
          "https://[YOUR_AZURE_DOMAIN].openai.azure.com",
      },
    },
  },
});

from langchain.chat_models import AzureChatOpenAI
helicone_headers = {
    "Helicone-Auth": f"Bearer {helicone_api_key}",
    "Helicone-OpenAI-Api-Base":
        "https://<model_name>.openai.azure.com/"
}
self.model = AzureChatOpenAI(
openai_api_base="https://oai.helicone.ai",
deployment_name="gpt-35-turbo",
openai_api_key=<AZURE_OPENAI_API_KEY>,
openai_api_version="2023-05-15",
openai_api_type="azure",
max_retries=max_retries,
headers=helicone_headers,
**kwargs,
)

  curl --request POST \
  --url https://oai.helicone.ai/openai/deployments/DEPLOYMENTNAME/chat/completions?api-version=[API_VERSION] \
  --header 'Helicone-Auth: Bearer [HELICONE_KEY]' \
  --header 'Helicone-OpenAI-Api-Base: https://[AZUREDOMAIN].openai.azure.com' \
  --header 'api-key: [AZURE_API_KEY]' \
  --header 'content-type: application/json' \
  --data '{
    "messages": [
      {
        "role": "user",
        "content": "Answer in one word"
      }],
    "max_tokens": 800,
    "temperature": 1,
    "model": "gpt-3.5-turbo-0613"
  }'

Getting Started

Integrations

Tracing

Prompt Engineering

AI Gateway

References