ChatCompletion->create_async

[asnagni]

Hi guys,
Do you have anywhere in the documentation an example showcasing the Chatcompletion with an asynchronus call passing a callback function?

  **ChatCompletion->create_async**

I couldn't find any example in the documentation.

The issue that I'm facing is that the AI agent will send me a reply in two or 3 parts (not in one shot). How can I make sure that I receive all data?

Thank you,
Stay safe

[pkusunjy]

Hi @asnagni ,

Perhaps you need to pass your own callback function to the stream parameter in create_async, and here is a tiny demo illustrating how it works.

// define your own callback function
auto callback = [](std::string data, intptr_t ptr) -> bool {
    std::cout << data << '\n';
    return true;
};

// pass it to stream
xxx.create_async(model, conversation, std::nullopt, std::nullopt, std::nullopt, callback);

The stream will return streaming messages with data: {somejsondata} and be closed by a data: [DONE] message.

Hope this may help you.

[asnagni]

Sorry for the delay and thank you for you answer, I do appreciate.

Stay safe