Getting Started with the LLM Chatbot Template
Welcome to webAI: Launch Your First LLM Chatbot!
This article will guide you through installing Navigator and Companion and help you launch your first LLM Chatbot within minutes!
If you have already downloaded and signed into Navigator, please skip to " Step 2 - Building Your LLM Chatbot" section.
Step 1 - Installing and Launching Navigator
For the best user experience, please ensure your device meets the necessary requirements here: Device Compatibility
-
Download the webAI Installer
This will provide access to the webAI Installer, our package manager. Once you've added the webAI Installer to your applications, launch it.
- Click on the Install Both button to install both Navigator and Companion.
-
Once Navigator has been installed, click on Launch.
- This will display an account creation form. Complete the required forms and submit!
Once you've successfully created your account and signed in, you can move onto the next step: Launching your LLM chatbot!
Step 2 - Building your LLM Chatbot
The fastest way to get an LLM Chatbot up and running is by using one of our templates available in the Featured Templates section inside Navigator. It allows you to utilize an existing, tried-and-true workflow that our webAI Experts have already crafted!
On Navigator:
-
Click on the LLM Chatbot template in the Featured Templates section and click Open.
This will direct you to your first project! Here you will see two elements already connected.
-
Now, all you need to do is click on the Run button in the upper right corner.
Here is a quick video that walks you through selecting a featured template, to running your LLM Chatbot!Note: since this is likely your first time running a model on your device using Navigator, it may take a few minutes to download and become active.
- Once the download is complete, your canvas will automatically switch to the Preview view, where you can begin interacting with your bot!
You have now successfully created your first local AI model on your device!
Step 3 (Optional) - Adjusting your LLM Chatbot's Settings
Within the Preview view, where you can interact directly with your LLM Chatbot inside Navigator, you also have the option to fine-tune your flow’s settings.
To make adjustments, click on the dropdown menu next to the Active button and select Stop.
Once your flow has been stopped, the settings on the left-hand side will become editable. You’ll notice that your preview status now displays as Disabled until you run your flow again.
For a detailed breakdown of each configuration option, refer to our LLM Element article.
Once you are satisfied with your chatbot's behavior, you're ready to take it to the next level by deploying it. You can now manage your deployments in the Apps view.
For more information on Deployments, visit our Deployments Overview documentation.
For more information on the different view options, refer to our Project Views article.
Quick Tips
- By default, the LLM Element uses Meta Lama 3.1 8B Instruct 4bit. You can adjust the model by clicking on the LLM Element and clicking on the Large Language Model Dropdown. When selecting a different LLM, please make sure the models you select are suitable for the available memory on your device.
-
Models marked with the following symbol are Gated models, meaning you will need to enter your Hugging Face API Key.
If you are using a gated model for the first time, you will need to make sure you have requested access to that model within Hugging Face and your request has been approved.
- For additional information on the LLM Element and the API Element's settings, you can visit our Element Registry.
If you're looking to take your LLM Chatbot to the next level, continue reading to learn about deploying a custom chatbot on Companion! Getting Started with Companion