Old Man
Old Man
It'd be nice if people stopped using discord...just sayin'
Same problem here
That didn't work for me. This did: "Use --net="host" in your docker run command, then localhost in your docker container will point to your docker host."
> Semi-related, but isn't k-quant the newer/better quantization method? I have found it confusing that ollama defaults to the non-K quants, but maybe I'm confused about which method is better....
> As a quick hack, rightclick the slider, select inspect, the browser element can be changed a higher maximum. > > This should work for anything ui related, like models...
> It's a little annoying that I still can't run the SOTA 35B model in the most popular webUI. I'm running it just fine. Let me know if you need...
> I couldn't get it running on Linux and a 7900xtx, tried both transformers and llamaCPP. I have it running on Linux and a 4090, llama.cpp through ooba. Good luck...
> > > It's a little annoying that I still can't run the SOTA 35B model in the most popular webUI. > > > > > > I'm running it...
> > > > > It's a little annoying that I still can't run the SOTA 35B model in the most popular webUI. > > > > > > >...