[baq]

I had only cuda installed and it took 2 ollama shell commands in WSL2 from quite literally 0 local LLM experience to running mixtral fast enough on a 1070 and 12700k. Go for it.