LLM

Our LLM package is a powerful tool that can generate text based on a prompt. You can use the GPT-3.5 or GPT-4o models from OpenAI to generate text, with more models from other providers coming soon.

​

Video Tutorial

Watch this video to see how to create a simple package using the LLM node.

​

How do I add it to my package?

To add the LLM node to your package, right click on the workspace grid and select Add Item. Then, select the LLM from the list of available nodes.

​

Configuring the LLM node

When you first add the LLM node to your package, you will see an Untitled node with gpt-3.5 selected as the model. Let’s change the name of the node to something more descriptive, and set the prompt.

​

Changing the name of the node

On the task panel on the right hand side of the screen, click on the “Untitled” text in the input field to change the name. As soon as you change the name, it will be saved automatically. For the purposes of this example, lets change our node name to Generate Joke. And now, here’s how our screen looks now that we’ve changed the name. Perfect! We’ve got a descriptive name for our node. Now let’s set the prompt!

​

Setting the prompt

To set the prompt, click on the node in the workspace, and on the right-hand panel, edit the Prompt field. Lets tell GPT to tell us a joke about cats. Now that we’ve set the prompt, there is one last step. We have to add a Finished node to tell packages how we want this output to be formatted. If we don’t add a finished node, we won’t be able to see our output. So lets do that!

​

Adding a Finished node

The finished node tells the app how to format the output from all the various nodes within a package. Right click on the workspace, and select Add Item. Then choose 🏁 Finished.

​

Changing the name

Just like with the LLM node, you can change the name of the finished node by clicking on the Untitled text in the input field on the right-hand panel. For this example, lets change the name of the finished node to All Done.

​

Including the results from the LLM node

To get the output from our Generate Joke node, we need to ”@ tag” the Generate Joke node in our output. To do that, head to the Template portion of the finished node configuration on the right side of the screen. Type @Generate Joke in the template field, and you’ll see a dropdown appear. Once you’ve selected the Generate Joke node, you’ll see a placeholder appear in the template field.

​

Run the package

Your first package using the LLM node is now ready to run! Hit the run package button to see the output. And there you have it! You’ve successfully created a simple package using the LLM node.

​

Configuration options

Here are all the different configuration options you can use with the LLM node.