Easy scripting with `node-red-contrib-custom-chatgpt` for beginners

[mighty mortar]

Are you looking for a way to easily incorporate the powerful capabilities of OpenAI's machine learning models into your workflows?

node-red-contrib-custom-chatgpt

Is a relatively new Node for Node-RED that enables easy integration with OpenAI machine learning models like ChatGPT. With this Node, you can leverage the power of OpenAI's natural language processing capabilities within your tools/workflows for dynamically generating code or other creative tasks.

Node-RED is a JavaScript based low-code programming tool that allows you to easily connect various devices and services together to create powerful tools/workflows. And now, with the addition of "node-red-contrib-custom-chatgpt," you can easily integrate OpenAI's natural language processing models into your workflows without needing any prior coding experience.

To get started, simply install Node-RED and the "node-red-contrib-custom-chatgpt" Node. Then, follow the instructions provided on the Github repository to configure the Node with your OpenAI API key and model parameters. From there, you can start using the Node to perform various tasks.

Even if you're new to Node-RED, don't worry - it's easy to use and requires no prior coding experience. Simply drag and drop nodes onto the canvas, connect them together, and configure them using the provided interface.

https://nodered.org/docs/getting-started/ will help you get Node-RED installed and running in just a few minutes.

And if you have any questions or need assistance, feel free to open an issue on the Github repository.

Github repository link: https://github.com/HaroldPetersInskipp/node-red-contrib-chatgpt

Screenshots:
https://raw.githubusercontent.com/HaroldPetersInskipp/node-red-contrib-chatgpt/main/examples/example.png

https://raw.githubusercontent.com/HaroldPetersInskipp/node-red-contrib-chatgpt/main/examples/example2.png

https://raw.githubusercontent.com/HaroldPetersInskipp/node-red-contrib-chatgpt/main/examples/example3.png

[mighty mortar]

Currently supports the following models:
text-davinci-edit-001,
text-davinci-003,
gpt-3.5-turbo,
As well as image creation via DALL·E models.

GPT-4 support coming soon.

[mighty mortar]

Preliminary GPT-4 support is now included. (Untested, I'm still on the waitlist)
The following optional inputs now have preliminary support as well - msg.suffix, msg.n, msg.temperature, msg.top_p, msg.presence_penalty, msg.frequency_penalty, and msg.echo

Easy scripting with node-red-contrib-custom-chatgpt for beginners

Don't be afraid to give it a try. No coding experience required, Node-RED is a low-code/no-code platform designed for both beginners and experienced developers.

[iron sun]

Looks like text-davinci-edit-001 and text-davinci-003 were removed

[mighty mortar]

Both models text-davinci-edit-001 and text-davinci-003 are now functional.

According to "logankilpatrick" Developer Advocate (OpenAI Staff) on the OpenAI community forums:
text-davinci-edit-001 is now back online, they have apologized for the delay, pointing out it was a misconfiguration on their side.

And it seems text-davinci-003 was always functional on my end. Sorry for any confusion this may have caused.

[iron sun]

https://community.openai.com/t/the-model-text-davinci-edit-001-does-not-exist/116144/28

Here's the link to the community post for reference

[mighty mortar]

Recent changes:

Added

• Included support for using case insensitive topics input with the msg.topic property.
• Included the ability to set the behavior Topic in the node's edit dialog. Once set-up the node can be controlled with as little as a single required message property msg.payload. Behavior can still be set dynamically by selecting read from msg.topic and setting the msg.topic to the desired behavior.

Changed

• Updated help information displayed in the info tab of the editor. (less verbose "details" section)
• Updated error message and removed unnecessary whitespace.
• Updated README.md formatting and usage information.
• Updated help information displayed in the info tab of the editor. (msg.max_tokens support is now documented)
• Updated the included example to demonstrate new features.

These new changes make it easier than ever to learn and use this node even if you are a beginner (no experience programing). After installing the node in Node-RED you can import the included example flow: https://github.com/HaroldPetersInskipp/node-red-contrib-chatgpt/blob/main/examples/chatgpt.json into the editor to assist with learning how it is used and some of its capabilities.

Importing and Exporting Flows in Node-RED

Flows can be imported and exported from the editor using their JSON format, making it very easy to share flows with others.

• Importing flows

The Import dialog can be used to import a flow by the following methods:

1. pasting in the flow JSON directly,
2. uploading a flow JSON file,
3. browsing the local flow library,
4.  browsing the example flows provided by installed nodes.

In all cases, the dialog offers the option to import the nodes into the current flow, or to create a new flow for them.

• Exporting flows

The Export dialog can be used to copy the flow JSON out of the editor by the following methods:

1. copying the JSON to the system clipboard,
2. downloading the JSON as a file,
3. saving it to the local flow library.

It can export either the selected nodes, the current flow (including its tab node) or the complete flow configuration.

It offers the option to export compact or formatted JSON. The compact option generates a single line of JSON with no whitespace. The formatted JSON option is formatted over multiple lines with full indentation - which can be easier to read.

For more information visit https://nodered.org/docs/user-guide/editor/workspace/import-export

[mighty mortar]

Recent updates:

Added

• Included additional example. generate-node-red-nodes.json demonstrates how to generate Node-RED nodes and import them directly into the editor automatically (proof of concept for Windows). OpenAI models will write the code for you based on simple parameters.
• Included support for using a proxy to access the OpenAI API. Set the BaseUrl property value to the proxy/vpn service's url that you would like to use.

Changed

• Updated help information displayed in the info tab of the editor. (improved formatting and documentation)
• Fixed properly setting msg.topic to lowercase characters.
• Removed unnecessary whitespace in node edit dialog.

Import the newly included example by updating the node or it can be viewed here: https://github.com/HaroldPetersInskipp/node-red-contrib-chatgpt/blob/main/examples/generate-node-red-nodes.json and then imported into your editor by following the above instructions.

https://raw.githubusercontent.com/HaroldPetersInskipp/node-red-contrib-chatgpt/main/examples/example5.png