I needed a way to catch up on the specific tech news that I need to know each week in podcast format so I created this pipeline using Hermes, Deepseek v4 Pro, and my Mac Studio - no frontier AI at all. I'm surprised by the quality. Local LLMs are the future, at least when the cost of the hardware comes back down or heavily quantized models become more accurate.
I needed a way to catch up on the specific tech news that I need to know each week in podcast format so I created this pipeline using Hermes, Deepseek v4 Pro, and my Mac Studio - no frontier AI at all. I'm surprised by the quality. Local LLMs are the future, at least when the cost of the hardware comes back down or heavily quantized models become more accurate.