Open Interpreter’s Local III is out today. We are building computer-controlling agents that work offline. This is our biggest step forward. - interpreter --local sets up fast, local LLMs. - We are hosting a free inference endpoint. - We are training our own model. ⬤

much, much more in this update. all local. read on:

Have you tried the new Fire Function V2 with it? It's supposed to be on "GPT-4o" levels at function calling.

woah no! if it's on ollama, we can test it with one command (interpreter --model ollama/firefunctionv2) but i think it's not yet. does it work through LM studio or Jan yet?

I think you found the perfect alignment for OI next milestones. local-first + free inferences is a flex, but training your own LAM is the finality we all need imo

feel the same

I still don't understand how to use this (I'm a marketing person instead of a coder/developer) Do I just download it on my Windows computer and then install it and start asking it to perform tasks for me in my offline environment? Not trying to be rude! Asking because I want to use it 👍👍

yes! you can run our one-click installers for Windows— if that doesn't work, try following the instructions at the top of this page:

@OpenInterpreter when is 01 light being shipped? I haven’t received any updates.

@OpenInterpreter we're on track to have it out this fall! you can build a device with the m5 stack right now: the device we're shipping will be more custom, higher quality speaker. working on a big 01 update + more details re: 🇺🇸 manufacturing later this month.

Woah. This is insane. Keep cooking!!

🫡 thanks seth!