I don't understand why I'm getting this error and none of my assistants are working anymore.

The error

pipeline-error-openai-llm-failed

means there was a failure in the OpenAI LLM processing pipeline, often due to issues like invalid API keys, model unavailability, or provider-side problems ([see documentation](https://docs.vapi.ai/calls/call-ended-reason)). To help you further, could you confirm if you have recently changed your OpenAI API key, model settings, or made any updates to your assistant configuration? Sources: - [Call Ended Reasons – Vapi Documentation](https://docs.vapi.ai/calls/call-ended-reason) - [Assistant Hooks – Vapi Documentation](https://docs.vapi.ai/assistants/assistant-hooks)

Hey! To help track down this issue, could you share: - The call ID This would really help us figure out what went wrong!

@User 4a0b7741-4b96-452e-8fe3-fda7102e0995 1aae8155-64b0-43a9-a229-337933f5b6a1 d671cb92-8997-441a-8e08-f965c19eb9d1 094722c2-db47-4f54-b9b5-fe891fb59657 8800e6af-b1e6-44f2-bc53-226514d806b7 bc803f40-57e3-465e-be78-e9c457f755fa

Hey, we have the same error! None of our Assistans working right now.

I don't have any answer. This problem has been going on for three days. But I thought it was my fault. I'm also waiting for a response from the VAPI SUPPORT department.

Hey Felix WIlliam! The issue was on our side, which has been fixed. It was related to our routing logic, which led it to default to the non tools OpenAI model. We have deployed the changes to both the weekly and daily channels, so if you are on either of them, it will work for you. I will suggest trying two to three test calls and let me know how it goes for you.

@Shubham Bajaj thank you. I noticed that are problems with you lastest "ChatGPT 4o(latest) Cluster" "An error occurred in this call: call.in-progress.error-pipeline-no-available-llm-model" you can verify at call id: 626d286b-c93a-42c5-9e87-90dfbebeefa3

Action Step: Check your assistant's model configuration to ensure a valid and available LLM (such as GPT-4o) is selected, and that your API keys are correct. If the error persists, it may be a provider-side issue—please contact [Vapi Support](support@vapi.ai) with your call ID for further investigation. Reference: [Call Ended Reasons – Vapi Documentation](https://docs.vapi.ai/calls/call-ended-reason) Source: - [Call Ended Reasons – Vapi Documentation](https://docs.vapi.ai/calls/call-ended-reason)

Hey Felix William, GPT 4.0 latest doesn't support Tools Calling. You can verify the same information in OpenAI Docs as well. I'll suggest you to change model to GPT 4.0, rest is your decision.