News for hackers. Stuff that matters.
Re:TinyTroupe, a new LLM-powered multiagent persona simulati... by simonw on Monday November 11, 2024 @12:12PM (#42108699)
It looks like this defaults to GPT-4o: https://github.com/microsoft/TinyTroupe/blob/7ae16568ad1c4de...

If you're going to try this out I would strongly recommend running it against GPT-4o mini instead. Mini is 16x cheaper and I'm confident the results you'll get out of it won't be 1/16th as good for this kind of experiment.

Re:TinyTroupe, a new LLM-powered multiagent persona simulati... by ttul on Monday November 11, 2024 @12:36PM (#42108869)
I suppose the Microsoft researchers default to 4o because the models are free in their environment…