getting no LLM found error
# support
m
how to chose Azure open ai llm as I am getting no LLM found error
v
Hi there! 👋 I see you're encountering a "no LLM found error" when trying to use Azure OpenAI. I can definitely help you with that! Based on the information I have, here's how you can configure Azure OpenAI LLM to resolve this error: ### 1. Set Up Azure OpenAI Credentials First, you need to configure your Azure OpenAI credentials in the Vapi dashboard: 1. Go to the Provider Keys dashboard tab at [dashboard.vapi.ai/keys](https://dashboard.vapi.ai/keys) 2. Add your Azure OpenAI credentials with the following information: - Set provider to "azure-openai" - Specify the model(s) you want to use (e.g., "gpt-4o-2024-08-06") - Set the correct Azure region for your deployment - Provide your Azure OpenAI endpoint and API key Here's an example of how to add these credentials via API:
Copy code
curl --location 'https://api.vapi.ai/credential' \
--header 'authorization: Bearer YOUR_API_KEY_HERE' \
--header 'content-type: application/json' \
--data '{
    "provider": "azure-openai",
    "models": [
        "gpt-4o-2024-08-06"
    ],
    "region": "YOUR_REGION",
    "openAIEndpoint": "YOUR_ENDPOINT",
    "openAIKey": "YOUR_API_KEY"
}'
### 2. Important Configuration Details - **Endpoint Format**: The endpoint should be the shorter resource endpoint, not the longer deployment endpoint. It should go up to
chat/completions?api-version=
without including the version number ([source](https://discord.com/channels/1211482211119796234/1211483291191083018/threads/1338649953102463038)). - **Model Name**: Ensure the deployment name matches exactly the model name provided in the VAPI menu (e.g., "gpt-4o-2024-08-06") ([source](https://discord.com/channels/1211482211119796234/1211483291191083018/threads/1339956981297057892)).
- **Region Availability**: Verify that your selected model is available in your configured region. For example, "gpt-4o-2024-08-06" might not be available in all regions ([source](https://discord.com/channels/1211482211119796234/1211483291191083018/threads/1343704609096601693)). ### 3. Resolve Credential Conflicts If you have both OpenAI and Azure OpenAI credentials configured, the system might favor regular OpenAI. To ensure Azure OpenAI is used, consider removing the regular OpenAI credential if you want to exclusively use Azure OpenAI ([source](https://discord.com/channels/1211482211119796234/1211483291191083018/threads/1342900897914556467)). ### 4. Specify Provider in Assistant Configuration When configuring your assistant, make sure to set the provider to "azure-openai":
Copy code
json
{
  "model": {
    "provider": "azure-openai",
    "model": "gpt-4o-2024-08-06"
  }
}
Is there anything specific about your setup that you'd like me to help troubleshoot further? Sources: - [Discord Thread on Azure OpenAI Configuration](https://discord.com/channels/1211482211119796234/1211483291191083018/threads/1338649953102463038) - [Discord Thread on Azure OpenAI Model Naming](https://discord.com/channels/1211482211119796234/1211483291191083018/threads/1339956981297057892) - [Discord Thread on Region Availability](https://discord.com/channels/1211482211119796234/1211483291191083018/threads/1343704609096601693) - [Discord Thread on Credential Conflicts](https://discord.com/channels/1211482211119796234/1211483291191083018/threads/1342900897914556467) - [Vapi Dashboard Provider Keys](https://dashboard.vapi.ai/keys)
k
@Manish was this guild helpful
s
checking if this is resolved/solved for you?
m
I am able to add credentials but it is not able to find the model
@Shubham Bajaj @Kings_big💫
s
Hey! To help track down this issue, could you share: - The call ID This would really help us figure out what went wrong!
m
aaa9b433-dd56-4f8b-b3e6-79739c953e7a
@Shubham Bajaj did you look into this
s
I am currently looking into this
m
Hi Kyle any update on this?
s
Not yet, still investigating. Thank you for your patience while I research this
c
Any Update on this?
s
This ticket has been marked as solved, and as a result, you will no longer receive any further responses. Kindly create a new support ticket to escalate your inquiry. 
2 Views