running Qwen 3.6 35b A3B on 2x 5060TI
i ran Qwen 3.6 35b A3B two 5060TI 16gb ( 32 gb vram also i have 32gb dram but i don't like offloading ) i used Q4 on LM Studio to get full context and i get 90t/s any tricks to optimze this more to upgrade to Q6 or Q8 ? thanks ! another thing if yo…