https://vapi.ai/ logo
Join Discord
Powered by
I cant make the trasferCall function to trigger in...
# support
f
a
HeyFelix WIlliam Thank you for creating the ticket! A member of our team will reach out to you shortly.
v
You can continue this conversation by mentioning me in the message.
@Felix WIlliam To trigger the 
transferCall
function from a custom LLM (Language Learning Model), you would typically need to have access to the source code where the function is defined. Since you've provided a GitHub link, I would first check the 
main.py
file in the repository to understand how the function is implemented and what parameters it requires. However, as an API support bot, I don't have the capability to directly access external URLs or GitHub repositories. If you can provide me with the function signature or details on how 
transferCall
is supposed to be used, I can guide you on how to trigger it. If 
transferCall
is a function that should be triggered through VAPI, you would normally need to send a request to the API with the necessary parameters. This could be done through an HTTP request or by using a specific SDK or library provided for interacting with VAPI. If you have the function signature or any specific details about the 
transferCall
function, please share them, and I'll do my best to assist you further. If you need to know how to trigger it from within a Python script, you would typically import the module containing the function and call it with the required arguments.
f
v
@Felix WIlliam Please ask support for more info on this.
f
@User can help ?
@Vapi Maybe there is a template for a custom LLM with a transferCall function inside, so we can build around it. The custom LLM must trigger the transferCall function at some point, using a phone number that is already defined in the assistant’s settings.
v
@Felix WIlliam It sounds like you're looking for a way to integrate a 
transferCall
function within a custom Language Learning Model (LLM) that can be triggered under certain conditions. If the 
transferCall
function is part of the assistant's capabilities and a 
forwardingPhoneNumber
is already set in the assistant's settings, you would typically define a condition within your LLM's logic to call this function when required. Here's a conceptual example of how you might structure this in pseudocode:
Copy code
python
# This is a simplified example and not actual code.
def handle_user_input(user_input):
    # Logic to determine if the call should be transferred
    if should_transfer_call(user_input):
        transfer_call_to_phone_number(forwardingPhoneNumber)

def should_transfer_call(user_input):
    # Analyze the user input and return True if the call should be transferred
    # This is where your custom logic will determine if a transfer is needed
    pass

def transfer_call_to_phone_number(phone_number):
    # This function would trigger the transferCall function
    # You would need to replace this with the actual function call
    transferCall(phone_number)
In this example, 
handle_user_input
would be the function that processes the user's input. If the logic determines that the call should be transferred, it calls 
transfer_call_to_phone_number
, which in turn would trigger the 
transferCall
function with the 
forwardingPhoneNumber
. To implement this in a real scenario, you would need to know the exact syntax and parameters required by the 
transferCall
function. If the 
transferCall
function is part of an API or a library you're using, you would need to refer to the documentation for the correct usage. If you can provide more details or if there's a specific condition under which the call should be transferred, I can give you more tailored guidance.
s
@Felix WIlliam Could you please provide the call ID that is encountering the failure?
@Felix WIlliam I apologize for not being able to respond to you earlier. After our conversation in another thread, I was included in additional discussions that required my immediate attention, which delayed my reply. I am available to connect over a call to assist you in resolving this matter. Please let me know a suitable time, excluding my already booked slots, or I can also block time on my calendar for our discussion.
@Felix WIlliam here you go with CustomLLM x Tools Calling according to your code https://codesandbox.io/p/devbox/gfwztp
f
Thank you. You CustomLLM x Tools Calling according to your code https://codesandbox.io/p/devbox/gfwztp IS working. Is very complex LLM ( i am not a programmer ) If you will have the time to shedule a meeting and help me understand some things wil be great. I can pay for help.
s
@Felix WIlliam No need to pay, can help you out for sure. You can join office hours today for your doubts https://discord.gg/8xsMTjEp?event=1335330727780286597
m
@Shubham Bajaj @Felix WIlliam I am now working on CustomLLM in FastAPI ( should be similar to Flask ) how can I call the default tool 
transferCall
from my customllm - 
/chat/completions
endpoint?
a
PreviousNext
Powered by