I'm very excited to launch this new plugin that supports OpenAI's latest GPT-4o model which you can read more about here.
This new plugin update also includes the ability to have GPT process images, which you can easily turn on and off with each new message via the 'Upload Image With Message' toggle. You can attach any TOP to the component and the second you press 'Generate Response' it will snag the current frame from the input TOP and pass it along to GPT along with your message.
Here's a link to my previous tutorial that shows how to get an API key and set up your project (new tutorial coming soon)
Here's another tutorial showing how to combine the plugin with other AI models to create a voice-activated talking 3D character that also follows the viewer's movement.
If you're using the vision feature don't forget to clear your chat history eventually, because it will keep the images encoded as text inside the component, which can take up a decent amount of space over time.
⚠️ Don't forget to include the TDAsyncIO.tox in your same project to allow the component to run⚠️
Phil Jungschlaeger
2025-01-02 19:30:40 +0000 UTCTorin Blankensmith
2024-12-15 18:11:44 +0000 UTCFilippo Mirandola
2024-12-14 17:44:32 +0000 UTCViola Liang
2024-11-25 15:34:06 +0000 UTCViola Liang
2024-11-24 19:49:28 +0000 UTCTorin Blankensmith
2024-11-24 19:34:52 +0000 UTCViola Liang
2024-11-24 19:08:14 +0000 UTCTorin Blankensmith
2024-11-24 18:24:12 +0000 UTCViola Liang
2024-11-24 16:51:44 +0000 UTCElrashid Ahmed
2024-08-27 09:00:58 +0000 UTC7337 Productions
2024-05-13 23:56:22 +0000 UTC