Wow, I can even chat about C code with that model with LM Studio on my Macbook at 200 tokens per seconds