New OpenAI features - Assistant API

Hey team ,

Is there any plans for new features/actions that incorporate the new features from OpenAI released yesterday?

2 Likes

We’d love our community’s input on this exact thing. If you have ideas, we’d love to hear them!

Looking forward to the inout tokens being adapted in order to make my content writing app more robust.

Do you mean the ability to send in larger amounts of input text?

1 Like

We’ve already incorporated the new model in our app (CopyCoached - Now on glide!) , however haven’t figured out how to get it to read an image yet as Glide turns images into URLs.

I was just thinking would the Vision API work direct in Glide somehow

Technically yes…but the problem I think is how it is passing images.

FYI we updated the OpenAI integration this week to support all the latest models as well as the ability to pass in images (vision mode) and even use JSON mode (I would be remiss if I didn’t mention that Glide AI already supports “vision” mode with the “Image to text” action).

Most of these updates can be found in the OpenAI “Complete chat” action. Let us know what you think!

(doc updates coming soon)

4 Likes

Fantastic :+1: - impressive that you have integrated these AI updates at such pace!

Looking forward to having a play around, just need to get my creative juices flowing (sipping a strong black coffee :coffee: )

1 Like

@J_Woosley Do you support plugs-in like AI PDF (for large pdf-files), keymate/search (for browsing the web) or webpilot for chating with urls? Or similar.

Sorry, no. The OpenAI API does not support browsing or plugins, so this would not be possible.

1 Like