6y56h56··on We fine-tuned Llama 405B on AMD GPUsThank you, I did not know there is an alternate route to achieve GPU usage. Will look into this.
6y56h56··on We fine-tuned Llama 405B on AMD GPUsI cannot get AMD ROCm running on my debian 12 system which is what I think is causing Ollama to use CPU instead of GPU. So I guess there is still a long way to go.