amd / RyzenAI-SW

MIT License
405 stars 65 forks source link

Demo consultation #96

Closed huzi-12 closed 4 months ago

huzi-12 commented 5 months ago

The demo shows LLama2 and OPT models. Do other LLM models support it? What is the maximum supported model parameter size? Is 13B supported? thank you

uday610 commented 5 months ago

Yes, some other LLM can be run similar to llama2 or OPT PyTorch flow. We will release flow/script to run more models soon.

ksjogo commented 4 months ago

@uday610 do you have a timeline for that? And are you running them from a fresh install?

uday610 commented 4 months ago

Yes they will be available with Ryzen AI 1.2 release should be in a month or so.

uday610 commented 4 months ago

closed