Yeah, but I don't really have the hardware. My personal PC is barely holding up for every day use and it only has integrated graphics.

i'm running an old 1080ti got it for 50 bucks

but.. skip that.. and go for one of these.. should fit in the closet just fine

I just had to rebuild my bathrooms, so...no.

How are you using it? If i try to talk with it over github copilot -> it hang on the first copilot-instrustion.md.

tried it over the vs code and still same results.

Check your token threshold. Assuming you use `LM Studio` you should be able to see the log as it runs/thinks.

i'm running it locally.. with the gpu i have i cant do more than 64k context otherwise myk tok/s drops a lot.. at 64k i can get 10tok/s at 256k it goes to 3..