Streaming Response With Custom Settings - ChatFly Documentation

API Endpoint

The API endpoint for obtaining a streaming response with custom setting is:

https://backend.chatfly.co/api/chat/get-streaming-response-with-custom-settings

To get a streaming response, you need to send a POST request to the API endpoint with a Params and JSON request. Here’s an example Params request:

{
    "type_qa": "String"
}

type_qa (string): type of question message.

And here’s an example JSON request:

{
  "bot_id": "3fa85f64-5717-4562-b3fc-2c963f66afa6",
  "message": "Hi",
  "chat_history": [
      {
        "sender_type": "user",
        "content": "Hi"
      },
      {
        "sender_type": "assistant",
        "content": "Hello! How can I assist you today?"
      }],
  "custom_prompt": "I want you to act as an assistant to a customer support agent by writing email responses...",
  "gpt_model_name": "gpt-3.5-turbo-16k",
  "temperature": 0,
  "rules": []
}

Example Request

Here are example commands to obtain a streaming response with custom settings using the ChatFly API:

curl -X 'POST' \
  'https://backend.chatfly.co/api/chat/get-streaming-response-with-custom-settings' \
  -H 'accept: application/json' \
  -H 'Content-Type: application/json' \
  -d '{
  "type_qa": "string",
  "bot_id": "xxxx-xxxx-xxxx-xxxx",
  "message": "Hi",
  "chat_history": [
      {
        "sender_type": "user",
        "content": "Hi"
      },
      {
        "sender_type": "assistant",
        "content": "Hello! How can I assist you today?"
      }],
  "custom_prompt": "I want you to act as an assistant to a customer support agent by writing email responses...",
  "gpt_model_name": "gpt-3.5-turbo-16k",
  "temperature": 0,
  "rules": []
}'

import requests

url = 'https://backend.chatfly.co/api/chat/get-streaming-response-with-custom-settings'
headers = {
    'accept': 'application/json',
    'Content-Type': 'application/json'
}
params = {
    'type_qa': 'string',
}
    
data = {
    'bot_id': 'xxxx-xxxx-xxxx-xxxx',
    'message': 'Hi',
    'chat_history': [
      {
        "sender_type": "user",
        "content": "Hi"
      },
      {
        "sender_type": "assistant",
        "content": "Hello! How can I assist you today?"
      }],
    'custom_prompt': 'I want you to act as an assistant to a customer support agent by writing email responses...',
    'gpt_model_name': 'gpt-3.5-turbo-16k',
    'temperature': 0,
    'rules': []
}

response = requests.post(url, json=data, headers=headers, params=params, stream=True)

if response.status_code == 200:
    print("Request successful!")

    # Process the streaming response as needed
    for chunk in response.iter_content(chunk_size=1024):
        if chunk:
            # Process each chunk of the streaming response
            print(chunk.decode('utf-8'), end="", flush=True)
else:
    print("Request failed with status code:", response.status_code)
    print(response.text)

const axios = require('axios');

const url = 'https://backend.chatfly.co/api/chat/get-streaming-response-with-custom-settings';
const headers = {
  'accept': 'application/json',
  'Content-Type': 'application/json'
};
const data = {
  'type_qa': 'string',
  'bot_id': 'xxxx-xxxx-xxxx-xxxx',
  'message': 'Hi',
  'chat_history': [
      {
        "sender_type": "user",
        "content": "Hi"
      },
      {
        "sender_type": "assistant",
        "content": "Hello! How can I assist you today?"
      }],
  'custom_prompt': 'I want you to act as an assistant to a customer support agent by writing email responses...',
  'gpt_model_name': 'gpt-3.5-turbo-16k',
  'temperature': 0,
  'rules': []
};

axios.post(url, data, { headers })
  .then(response => {
    console.log('Response Status Code:', response.status);
    console.log('Response Content:', response.data);
  })
  .catch(error => {
    console.error('Error:', error);
  });

Response Body

The streaming response is as follows:

*Answer:* Hello! How can I assist you today? 

*Suggest Questions:*
1. What specific information are you looking for?
2. Is there a particular department or service you need assistance with?
3. Can you provide more details about your inquiry or issue?
4. How can I help make your experience better?