skinny.

Latest / Misar.Blog Podcast

How to Run Llama 4 Scout on Mac Studio M4 Max

Step-by-step guide to run Llama 4 Scout locally on a Mac Studio M4 Max: which memory config you need (64GB vs 128GB), MLX and Ollama commands, realistic… From the article "How to Run Llama 4 Scout on Mac Studio M4 Max" by Synor, published on Misar.Blog.

In this episode:

0:00 Llama 4 Scout Runs Well on a Mac Studio M4 Max Only When the Device Has 64 GB

2:08 Scout Is a 109-B Parameter Mixture-Of-Experts Model

3:57 At 4-Bit Quantization Scout Requires About 55

5:43 The 4-Bit Scout Model’s Weights Alone (~57 GB) Exceed the GPU-Usable Portion

6:13 On a 64-GB M4 Max

6:36 On a 128-GB M4 Max

7:01 MLX…

The skinny

The skinny isn't ready yet — notes appear once the transcript is processed.