Bruce MacDonald
Bruce MacDonald
Resolving this to unify with #1910, which will be solved by #3080
related #2805
Hi @Vaayne thanks for the contribution, I've been working on this for the past couple of days too (my bad, should have assigned myself the issue). If it's ok with...
Hi @J0hnny007, thanks for opening the issue. Ollama only supports the Metal GPU API on Macs right now. AMD GPUs won't work.
Good call @cjhetzle, this is no longer an issue as we have since updated how we run models.
Hi @k3341095, which model is this and what are the resources allocated to the container?
@iliabaranov would you be able to share the output of your `nivida-smi` command?
Hi @oliverbob, this seems like a good case for fine-tuning or a different model. Before going with the fune-tuning approach I'd encourage you to try `dophin-mixtral` or something similar. Addressing...
Thanks for the contribution @purificant, we are actually targeting Go 1.20 intentionally at the moment for compatibility.