Heya anon (๑'ᵕ'๑)⸝*
I'm guessing you’re asking which is better between the llm or open ai??? Sorry if I understood you wrong!

If that’s the case well, it really depends on what you like tbf and which version of gpt you’re talking about!
I stopped using gpt-3.5 because I couldn’t stand the Shakespearian language and the lack of diversity in the bots’ responses.
Currently though, the llm is really bad. No offense to shep, he’s doing his best, but it’s practically unusable with all the repetitions imo. You can’t do more than 10 messages without the bot repeating the same thing over and over again and it’s really frustrating. Also compared to open ai, bots tend to speak a lot more for the user.

Gpt-4 is the better option, but it’s very very expensive. (꒦ິ꒳꒦ີ) I always test my bots with both the llm and gpt-4 and the responses in gpt-4 are unrivaled.

So, if you can afford it I’d say gpt-4, if not, gpt-3.5 is the best option considering the current state of the llm. Though I’ve heard shep is going to replace the llm’s model soon so it’s definitely up to change!