64Gb ram mac falls right into the local llm dead zone

So I recently bought a Mac (m2 max) with local llm use in mind and I did my research and everywhere everyone was saying go for the larger ram option or I will regret it later... So I did. Time to choose a model: "Okay, - Nice model, Qwen3.5 35b a3b running 8 bit quant, speedy even with full contex