I have created a model deployment in Azure with the name gpt-4o-2024-08-06. I have entered the details in the Azure OpenAI section of the Provider Credentials, and I get two green ticks. When I test the agent on the dashboard I get an error. It appears that the system is unable to connect to Azure OpenAI.

@jackg I'm glad to hear that you've set up your model deployment and have entered the details in the Provider Credentials section. The green ticks indicate that your credentials were accepted, but since you're experiencing an error when testing the agent, we should check a few things: 1. **Model Deployment Name**: Ensure that the model deployment name

gpt-4o-2024-08-06

is correctly specified in the configuration and that the model is deployed and running on Azure. 2. **Network Configuration**: Verify that there are no network issues or restrictions that might be preventing the connection from being established. 3. **API Endpoint**: Confirm that the API endpoint URL is correct and matches the one provided by Azure for your deployment. 4. **API Key**: Double-check that the API key is correct and has the necessary permissions to access the model. 5. **Error Details**: Look at the specific error message you're receiving for more clues. It can often point you in the right direction for troubleshooting. If you've checked all of the above and the issue persists, it might be helpful to provide me with the specific error message you're seeing, so I can assist you further. If the error message contains sensitive information, please redact it before sharing. If you're unsure about any of the details or how to proceed, please ask support for more info on this.

These are the logs from the failed calls: Voice cached: Hello how can i help? Assistant speech started Assistant speech stopped User speech possibly starting User speech started User speech possibly stopping Endpointing timeout 100ms (rule:

heuristic

) Transcriber output: Hello. Model request started

call_id?

Hi @Sahil call_id is 61938519-dbb3-4e7f-805b-8edba42487d5

Hey @jackg can you check again your deployed LLM model name, because it clearly says model name not found.

logs

🟡 10:40:43:085 Live Call Manager Ending With Error...(pipeline-no-available-model) 🟡 10:40:43:084 No Available Model For { "provider": "openai", "model": "gpt-4o", "semanticCachingEnabled": false, "fallbackModels": [], "temperature": 0.7, "maxTokens": 250, "functions": [], "tools": [], "toolIds": [], "messages": [ { "role": "system", "content": "You are a helpful assistant." } ], "numFastTurns": 0, "emotionRecognitionEnabled": false } 🟡 10:40:43:081 No model found to process LLM request

Hi @Shubham Bajaj the setup is: deployment name: gpt-4o-2024-08-06 model: gpt-4o endpoint url: https://MYNAME.openai.azure.com (full URL is https://MYNAME.openai.azure.com/openai/deployments/gpt-4o-2024-08-06/chat/completions?api-version=2024-02-15-preview) Is that correct? Thanks.

Hey @jackg the url is supposed to be

https://your name.openai.azure.com/openai

. I don't think the full URL is required but not sure of your native environment. Also I suggest testing your API url following this https://learn.microsoft.com/en-us/azure/api-management/azure-openai-api-from-specification#test-the-azure-openai-api

@Shubham Bajaj https://your name.openai.azure.com/openai does not get green ticks, so cannot connect. https://your name.openai.azure.com/ with the same key gets green ticks but does not work.

Hey @jackg can you try to join today's discord OH for live Vapi Tech Support.

what is the right format of setting azureopeni credentials..

Please create a new support ticket for your issue as this ticket has been solved and does not come under SLA, so you won't be getting any responses for the