Thread

v0.3.6 == vapourware

"With 64GB of RAM, you're in for a treat! You can explore a variety of language models, but the real fun comes from experimenting and finding the right fit for your projects. It's less about the specs and more about the creativity you bring to the table! 🚀✨ #AI #LanguageModels"

Most likely 7B or 14B. Mainly due to low CPU and no GPU. 64GB could extend into 32B models assuming you also need ram for all other services. But processing power is main issue with regard to mid to big size LLMs.

Up to 30B parameter should fit just fine, though it will still be bottlenecked by GPU, so only expect a few words per second.

Not until the next release

Awesome. Looking forward to it. 🍵

Whennnnnn? I set a goal to release 10 start9 apps this year and stopped at 3 because there is no path to users because I was told not to develop for 3.5 BC 3.6 is coming and thos was in November. Can y'all maybe update the community on what the hold up is because it's really disheartening to have put on hours of work to develop apps for a platform and have 0 users because of a blocker we have no control or transparency on. I don't want to switch to just developing stuff for hosting straight on proxmox BC for me it only matters if my boomer parents could possibly run my apps.

We have update and developer calls every Monday and Thursday from 10am-1pm MT. GitHub next/major branch is being updated daily. We've decided to move directly toStartOS 0.4.0 which is currently in Alpha testing and working great with many packages that use the new SDK. It's alove and working well, get in there! Maybe see you on the call Monday?