Chatbot Ai Settings

Jump to: Pusher Configuration Settings · Reverb Configuration Settings · Chatbot and Retrieval Settings · OCR & Tiktoken Settings · Video & Sitemap Settings · OpenAI API Settings · General Settings · Embedding Settings · GeoIP & Device Detection Settings

Pusher Configuration Settings

From this page, the admin can configure all Pusher-related settings. These settings ensure real-time communication features—such as instant chatbot responses and live updates—work seamlessly across the platform.

Screenshot

Pusher Configuration Settings page

Reverb Configuration Settings

This guide provides detailed, step-by-step instructions for configuring Reverb as your broadcasting driver. Reverb is Laravel's built-in WebSocket server, offering real-time communication without relying on third-party services like Pusher (though Pusher remains an optional alternative).

Screenshot

Prerequisites:

Ensure your script is installed and running on a Laravel-compatible server. You have administrative access to the application and server. For production, ensure your server supports running background processes (e.g., via Supervisor or systemd). Configuration Steps:

Step 1: Log In as Admin

Open your browser and navigate to the admin login page.
Enter your admin credentials (username/email and password) and click Login.

Step 2: Navigate to Pusher Settings

After logging in, Click Chatbot & AI Settings to access the broadcasting configuration options.

Step 3: Select the Broadcasting Driver

In the pusher settings page, locate the Broadcasting Driver.
Select Reverb from the options available.

Step 4: Fill in the Reverb Settings Fields

Reverb App ID: A unique identifier for your Reverb application. You can generate a random numeric or alphanumeric value (e.g., "290293"). 
Example App ID: 123456.

Reverb App Key: A public key for client-side authentication. Generate a secure, random string (e.g., "ikbkhfz3Jotfkin4koy"). Use a password generator. 
Example App Key: X7k9pLm2nQ8rStUvWxYzXilancerdEfGhIjKlMnOpQr12.

Reverb App Secret: A private key for server-side authentication. Generate a secure, random string (e.g., "fcfn4mtrI33tz0mwaewl"). 
Example App Secret: PqRsTuVwXyZXilancerDeFgHiJkLmNo12.

Reverb Host: The hostname or IP where the Reverb server runs. Use localhost for local development or your domain (e.g., your-domain.com) for production.

Reverb Port: The port for the Reverb server (default is 8080). Choose an open, unused port.

Reverb Scheme: Use http for local testing or https for secure production environments.
Important Notes on App ID and App Key:

You can generate random values for App ID, App Key, and App Secret as they are user-defined in Reverb. They do not need to be obtained from a third-party service. 
Use tools like a password generator or PHP's bin2hex(random_bytes(16)) for secure values.
Ensure the values are unique, secure, and consistent between the admin settings and server configuration to avoid authentication errors.

After filling in the fields, click the Update button to save changes. The .env file will be updated automatically with these settings.

Step 5: Set Up and Start the Reverb Server

Access your server via SSH or hosting control panel.

Navigate to the project root directory (e.g., cd /path/to/botmerze/core/).

Start the Reverb server:
For testing: Run php artisan reverb:start in the terminal (foreground mode).
For production: Use a process manager like Supervisor. 
Example config (/etc/supervisor/conf.d/reverb.conf):
[program:reverb]
command=php /path/to/botmerze/core/artisan reverb:start --host=0.0.0.0 --port=8080
directory=/path/to/botmerze/core
autostart=true
autorestart=true
user=www-data
numprocs=1
redirect_stderr=true
stdout_logfile=/path/to/botmerze/storage/logs/reverb.log
Apply changes with sudo supervisorctl reload.
Verify the server is running: Check logs or use telnet localhost 8080 (adjust host/port as needed).

Step 6: Test and Go Live

Return to the app and test the chatbot (e.g., send a message).

If configured correctly, real-time updates will work.Monitor logs (storage/logs/laravel.log and Reverb logs) for errors.

For further assistance, refer to the Laravel Reverb documentation at https://laravel.com/docs/reverb or contact our support.

Chatbot and Retrieval Settings

This page allows the admin to manage settings related to the chatbot’s retrieval and performance. Here, you can configure options such as chatbot enhancement, maximum retrieved knowledge length, session rate limits, and knowledge chunk limits to optimize chatbot efficiency and accuracy.

Screenshot

Chatbot and Retrieval Settings Page

OCR & Tiktoken Settings

This page allows the admin to configure OCR and TikToken management settings for the AI system:

Tesseract OCR Settings:

Executable Path: Set the full path to the Tesseract OCR executable on your server.
Purpose: Tesseract extracts text from image files. Ensure it is installed and the path is correct.
Installation Guide: Provides common paths for Ubuntu/Debian, CentOS/RHEL, macOS, and Windows.

Tiktoken Settings:

Cache Directory: Specify where the tiktoken vocabulary cache files are stored.
Default Encoder: Choose the token encoder (e.g., cl100k_base for GPT-4 and GPT-3.5-turbo).
Purpose: Tiktoken counts tokens in text to manage API costs and stay within model limits.

Admins can update both OCR and Tiktoken settings from this page to optimize AI performance and cost efficiency.

Screenshot

OCR & Tiktoken Settings Page

Video & Sitemap Settings

Screenshot

Video & Sitemap Settings

Install yt-dlp to /usr/local/bin/yt-dlp

1. Connect to your VPS

ssh user@your-server-ip

2. Install required dependencies

Make sure curl is installed.

sudo apt update
sudo apt install curl -y

3. Download yt-dlp to /usr/local/bin

sudo curl -L https://github.com/yt-dlp/yt-dlp/releases/latest/download/yt-dlp \
-o /usr/local/bin/yt-dlp

4. Make it executable

sudo chmod a+rx /usr/local/bin/yt-dlp

5. Verify installation

yt-dlp --version

/usr/local/bin/yt-dlp --version

Example output:

2026.02.25

Test Download

yt-dlp https://www.youtube.com/watch?v=dQw4w9WgXcQ

Optional (Recommended) Install ffmpeg

sudo apt install ffmpeg -y

Verify:

ffmpeg -version

Update yt-dlp later

sudo yt-dlp -U

Check installation path

which yt-dlp

Expected output:

/usr/local/bin/yt-dlp

OpenAI API Settings

From this page, the admin can configure the OpenAI API settings.

Screenshot

General Settings

This page lets the admin configure key general settings for the Chatbot and AI system. Here, you can manage options like the default AI model, temperature level, system prompt caching, and whether clients are allowed to use their own API keys.

Screenshot

Chatbot & AI General Settings

Embedding Settings

This page allows the admin to configure embedding-related settings for the AI system. You can manage options such as:

Embedding Model: Fixed to text-embedding-3-small.
Tokens Per Embedding Chunk: Number of tokens per knowledge chunk (128–1024).
Token Overlap: Overlapping tokens between chunks (0–500).
Minimum & Maximum Chunk Size: Define size limits for knowledge chunks.
Max Embedding Tokens: Maximum tokens allowed for a single document embedding (1000–16384).
Default Knowledge Doc/File Size: Maximum file size for uploaded knowledge files in KB.

These settings ensure efficient knowledge embedding and optimal AI performance.

Screenshot

GeoIP & Device Detection Settings

This page allows the admin to configure GeoIP location and device detection settings.

Screenshot

GeoIP & Device Detection Settings Page

Still stuck?

Our support team is ready to help you get set up.

Get support