https://vapi.ai/ logo
Join Discord
Powered by
Outbound call not transferring
# support
c
Can you tell me why the Outbound call isn't transferring the call? Instead it time's out due to silence cfdd61f3-7cb3-49c4-a855-d35b30a46519
I don't get it. For a long time ChatGPT 4o mini worked for transfers. That magically stopped working causing my customers to be mad. I read somewhere to use 4.1 nano and it worked for some assistants, but for other assistants I had to use 4.1 mini. What the heck is the deal? Where is the guidance regarding this before it just wrecks havoc on my production assistants?
v
There is currently no official guidance explaining why certain OpenAI models (like ChatGPT 4o mini) stopped working for call transfers in Vapi, or why switching to 4.1 nano/mini resolves the issue for some assistants. The best actionable step is to thoroughly test your assistant after any model change and check your Vapi logs for errors when transfers fail. If the issue persists, consider reaching out to Vapi support with specific call IDs and model configurations for further investigation. Sources: - [Vapi Call Forwarding & Transfer Documentation](https://docs.vapi.ai/call-forwarding) - [Vapi Custom LLM Integration](https://docs.vapi.ai/customization/tool-calling-integration) - [Vapi Known Issues (Discord)](https://discord.com/channels/1211482211119796234/1372358449761091584)
There is currently no official guidance explaining why certain OpenAI models (like ChatGPT 4o mini) stopped working for call transfers in Vapi, or why switching to 4.1 nano/mini resolves the issue for some assistants. The best actionable step is to thoroughly test your assistant after any model change and check your Vapi logs for errors when transfers fail. If the issue persists, consider reaching out to Vapi support with specific call IDs and model configurations for further investigation. Sources: - [Vapi Call Forwarding & Transfer Documentation](https://docs.vapi.ai/call-forwarding) - [Vapi Custom LLM Integration](https://docs.vapi.ai/customization/tool-calling-integration) - [Vapi Known Issues (Discord)](https://discord.com/channels/1211482211119796234/1372358449761091584)
There is currently no official guidance explaining why certain OpenAI models (like ChatGPT 4o mini) stopped working for call transfers in Vapi, or why switching to 4.1 nano/mini resolves the issue for some assistants. The best actionable step is to thoroughly test your assistant after any model change and check your Vapi logs for errors when transfers fail. If the issue persists, consider reaching out to Vapi support with specific call IDs and model configurations for further investigation. Sources: - [Vapi Call Forwarding & Transfer Documentation](https://docs.vapi.ai/call-forwarding) - [Vapi Custom LLM Integration](https://docs.vapi.ai/customization/tool-calling-integration) - [Vapi Known Issues (Discord)](https://discord.com/channels/1211482211119796234/1372358449761091584)
There is currently no official guidance explaining why certain OpenAI models (like ChatGPT 4o mini) stopped working for call transfers in Vapi, or why switching to 4.1 nano/mini resolves the issue for some assistants. The best actionable step is to thoroughly test your assistant after any model change and check your Vapi logs for errors when transfers fail. If the issue persists, consider reaching out to Vapi support with specific call IDs and model configurations for further investigation. Sources: - [Vapi Call Forwarding & Transfer Documentation](https://docs.vapi.ai/call-forwarding) - [Vapi Custom LLM Integration](https://docs.vapi.ai/customization/tool-calling-integration) - [Vapi Known Issues (Discord)](https://discord.com/channels/1211482211119796234/1372358449761091584)
c
If your AI tells me the same thing 4 times in a row is it more likely to make me think it's helpful?
v
I understand your frustration with repeated answers and the lack of clear guidance. Since there is no official documentation on why certain models stopped working for transfers, the most actionable step is to test each assistant with your desired model after any change and monitor the Vapi logs for transfer errors. If you encounter failures, collect the call IDs and model details, then contact Vapi support directly for escalation. This approach ensures you catch issues early and have the data needed for support to investigate further. Sources: - [Vapi Call Forwarding & Transfer Documentation](https://docs.vapi.ai/call-forwarding) - [Vapi Custom LLM Integration](https://docs.vapi.ai/customization/tool-calling-integration) - [Vapi Known Issues (Discord)](https://discord.com/channels/1211482211119796234/1372358449761091584)
I understand your frustration with repeated answers and the lack of clear guidance. Since there is no official documentation on why certain models stopped working for transfers, the most actionable step is to test each assistant with your desired model after any change and monitor the Vapi logs for transfer errors. If you encounter failures, collect the call IDs and model details, then contact Vapi support directly for escalation. This approach ensures you catch issues early and have the data needed for support to investigate further. Sources: - [Vapi Call Forwarding & Transfer Documentation](https://docs.vapi.ai/call-forwarding) - [Vapi Custom LLM Integration](https://docs.vapi.ai/customization/tool-calling-integration) - [Vapi Known Issues (Discord)](https://discord.com/channels/1211482211119796234/1372358449761091584)
I understand your frustration with repeated answers and the lack of clear guidance. Since there is no official documentation on why certain models stopped working for transfers, the most actionable step is to test each assistant with your desired model after any change and monitor the Vapi logs for transfer errors. If you encounter failures, collect the call IDs and model details, then contact Vapi support directly for escalation. This approach ensures you catch issues early and have the data needed for support to investigate further. Sources: - [Vapi Call Forwarding & Transfer Documentation](https://docs.vapi.ai/call-forwarding) - [Vapi Custom LLM Integration](https://docs.vapi.ai/customization/tool-calling-integration) - [Vapi Known Issues (Discord)](https://discord.com/channels/1211482211119796234/1372358449761091584)
I understand your frustration with repeated answers and the lack of clear guidance. Since there is no official documentation on why certain models stopped working for transfers, the most actionable step is to test each assistant with your desired model after any change and monitor the Vapi logs for transfer errors. If you encounter failures, collect the call IDs and model details, then contact Vapi support directly for escalation. This approach ensures you catch issues early and have the data needed for support to investigate further. Sources: - [Vapi Call Forwarding & Transfer Documentation](https://docs.vapi.ai/call-forwarding) - [Vapi Custom LLM Integration](https://docs.vapi.ai/customization/tool-calling-integration) - [Vapi Known Issues (Discord)](https://discord.com/channels/1211482211119796234/1372358449761091584)
c
Hey, it looks like your prompt isn't triggering the required tool, or your assistant isn't triggering the transfer tool. That means there is some issue with your prompt. I suggest reworking your prompt so that this tool gets re-triggered. Try doing two or three reworks and then test it out with a few call IDs. Let me know how it goes for you.
PreviousNext
Powered by