Method requestAsync

src/plugins/chat/chatmanager.cpp:267–286

For chatting: Using user-defined models

//For chatting: Using user-defined models
void ChatManager::requestAsync(const QString &prompt)
{
    if (!chatLLM || chatLLM->modelState() == AbstractLLM::Busy)
        return;

    answerFlag++;
    startReceiving();
    QtConcurrent::run([=]() {
        auto processedText = promptPreProcessing(prompt);
        if (processedText.isEmpty())
            return;
        auto c = chatLLM->getCurrentConversation();
        c->addUserData(processedText);
        QJsonObject obj = chatLLM->create(*c);
        if (isConnectToNetWork())
            obj.insert("command", "online_search_v1");   // only worked on CodeGeeX llm
        if (isRunning)   // incase stop generate before prompt preprocessed
            emit sendSyncRequest(obj);
    });
}

//For quick processing of special requests
QString ChatManager::requestSync(const QString &prompt)