Trying two other Deepseek R1 Distilled qwen 14B models with 1 million token one in MLX and one in GGUF formats to compare on LM Studio for my 14" Macbook Pro M4 Pro 14C/20GPU, 48GB ram, 1TB laptop - saving LLM models' downloads to external JEYI 2464Pro USB4 NVMe + 2TB Kingston KC3000 NVMe 😎