Carmine Paolino

But all of the other ones, you have to send the entire conversation.

Yeah.

So even for me, actually, it was really cool to have that feature because I wanted to test a different model while being in the same conversation.

So when I was selecting the models that I wanted to use,

I started to chat maybe with one model and then I was not really like pleased with their output.

So I was like, why can't I just use the same chat, but with a different model?

And that's how also that feature came about.

And in terms of token usage, I mean, you can also switch around between models.

So for example, if you exhaust like the tokens for a specific model, then you can switch around to another one that is perhaps a bigger context window and then you're going to be fine.

So you can just put it in your gem file, bundle install.

And then we also have a Rails integration and some generators that are going to help you to connect it all together.

No, no.

Wow.

That's right, yeah.

So basically, the Rails integration is built on top of the callbacks.

So every time we have a new message or there's a new chunk, we actually pass it to the Rails integration.

So it's using RubyLM features inside RubyLM in order to build the Rails integration.

And that allows us to have a very, very short amount of code for the Rails integration.

And on top of that, we have all the migration generators so that you can create your models that have all the different fields that we require.

And then in the end, in your controllers, you just have an access, let's say chat, if it's a chat or access message, if it's a message.