What is Flowise AI?
Flowwise AI is a tool that makes it incredibly easy to create powerful AI applications without writing a single line of code. It is open source, free to use, and was developed using the extremely powerful LangChain framework. If you are unfamiliar with LangChain, it is a development framework that allows developers to build AI-powered applications and has quickly grown in popularity.
And Flowwise makes all the features of LangChain available via an easy-to-use drag-and-drop user interface. So instead of having to write complicated code, all you have to do is drag-and-drop these nodes onto a canvas in order to create some complex AI applications.
Flowwise has become my favorite platform for creating AI chatbots and agents for my clients, and I can tell you from experience that it's possible to create some impressive AI applications using Flowwise. It's also easy to integrate Flowwise into our websites or applications using either the embedding or API endpoints. Let's start by setting up Flowwise locally.
Overview of Local Installation Options
We will have a look at deploying Flowwise to the cloud in this series, but it's always a good idea to start off with a local instance for development and testing. Setup is extremely easy, and we can see on the Flowwise website that it basically involves calling two commands in the terminal.
This will install Flowwise globally on our machines, which I'm personally not a fan of, so I will also show you two more methods for installing Flowwise in a specific folder. And if you stick around, I will also show you how you can use a simple script that I wrote specifically for this series to quickly install and update Flowwise.
Method 1: Global Installation with NPM
In order to use Flowwise, you do need Node.js installed, so go over to nodejs.org and install the LTS version of Node. Then open up your command prompt or terminal, and then enter the command npm install -g flowwise, and press enter. This will install Flowwise globally on your machine.
If you would like to install Flowwise in a specific folder instead, then skip ahead to the next method. For the majority of you, this method will be perfectly fine. After installation has completed, we can start Flowwise by executing npx flowwise start. This will start a local server, and we can access Flowwise by going to localhost:3000, and you should see a dashboard like this. So whenever you want to start Flowwise, open your terminal, enter npx flowwise start, and you can then access Flowwise by going to localhost:3000.
In order to update your Flowwise instance, you can simply run npm update -g flowwise, and this will update your Flowwise instance to the latest release. And if you ever wanted to uninstall Flowwise, you could simply run this command.
Method 2: Folder-Specific Installation with Git
Now for the majority of you, this installation process will work just fine, but if you wanted to install Flowwise into a specific folder, we can do that too. Installing Flowwise into its own folder actually opens up a few more advanced features, which we will have a look at during the course of this series. This is not too complicated, but it does involve a few steps, and after this I will show you how you can download a single script that will run all of these steps for you in one single go.
For this method, you do need get installed for your operating system. So download get from this website and install it. Then go over to the Flowwise repository. I will leave a link in the description of this video, or you can access it from the Flowwise AI website by clicking on GitHub. First, click on code, then copy this URL, then create a new folder on your machine, and open up the command prompt. In Windows, we can do that by clicking on this address bar and entering CMD. Then enter get clone and paste in that URL, and this will create a Flowwise subfolder, and if we open this up, we can see all the code that makes up Flowwise.
Then open the command prompt in this folder, so I'll simply enter cd Flowwise, and Enter. Now we need to install the dependencies for the Flowwise project, and we can do that by running yarn install. If you run this and get an error message, something in the lines of yarn is not available, you can easily install yarn by running npm install --global yarn, and you should then be able to run yarn install. This could take a few minutes to complete.
After installing the dependencies, we just have to run one more command, and that is yarn build, and we can now finally start Flowwise by running yarn start, and this will spin up the local server, which we can access from localhost:3000, and if I open up localhost:3000, I'm able to access Flowwise.
Method 3: Simplified Installation with a Custom Script
Now let me show you the simplest way to execute all of these previous steps in one single go. Go over to this repository, which I've created specifically for this series, and under branches, select executables. Go to code, and click on download zip. You can now extract the contents of that zip file anywhere on your machine, and you will be presented with two files - a run.bat file, and a run.shell file. If you're using Windows, you can double click on run.bat, and for Linux and Mac, you can click on run.shell. You might see a pop-up like this. Simply click on more info, and then click on run anyway.
The first time you run this, it will take a few minutes to complete, because it's installing all the dependencies for the first time, but if you ever want to use Flowwise again, simply double click this file, and the startup will be way faster, and if you want to ensure that you have the latest version of Flowwise installed, all you have to do is delete the Flowwise folder, and then execute run again. This will not delete any of your existing flows, and will simply download the latest project files and dependencies. The script will automatically open Flowwise in the browser, and now we can get started using Flowwise.
UI Tour: Dashboard & Marketplace
Before we create our first Flowwise application, let's have a look at the user interface. This is the chat flows dashboard, and this is where you can see all your existing Flowwise flows. You are able to search through your existing flows, and you can change your view from this grid view to a list view over here, and we can create a new chat flow by clicking add new, and this will bring up this canvas where we can select from all the available nodes, and we can view settings about this project, and we can click on this chat bubble to interact with our flow, but we will get back to that in a second. Let's go back to our dashboard, and let's have a look at a few other options in this menu.
In marketplace, we can view a list of pre-created templates that we can copy to get started, and under tools we can see a list of pre-created tools that we can plug into our applications. The marketplace is a fantastic way to learn Flowwise, and if we want to use one of these templates, all we have to do is click on the template, and then click on use template. We can now simply save this template, we can give it a name, I'll just call it demo, and then save it.
If we now go back to our dashboard, we can see this demo flow in our list of chat flows. We can delete this chat flow by clicking on it, and then within settings we can click on delete chat flow, and then delete.
Let's have a look at tools. We will go into this in detail during the series, but we are able to assign custom tools to our AI applications, and these tools could be anything, like a tool to fetch the current weather, or the current stock prices. We will have a look at creating these tools in the series.
We can use Flowwise to create, manage, and integrate OpenAI assistance, and this is also something we will have a look at in detail in the series.
Let's have a look at credentials. This is where we can manage all the credentials for our project, and as you'll see in the series, we will use several credentials, like the OpenAI API keys, and perhaps connections to other third parties as well, and we can add credentials by clicking on add credential. We can then select the provider that we would like to connect to, and that will store the credential over here, which we can reuse throughout our flows.
We can also use variables to set environment variables and secrets as well. Then lastly, we can create API keys, and this will allow us to call our Flowwise applications using API endpoints. This is also something that we will have a look at during the course of the series.
We are now ready to create our very first AI application using Flowwise.
