Hacker News Reader

Compute? Inference doesn't only need memory bandwidth. You need to actually do work with the memory you're loading which needs compute power. Which needs more electricity, which needs more cooling, which isn't practical for something as thin as a MBP.

View on HN

MacBook Pro has plenty of compute for local LLMs to be usable. I'm getting up to ~150 tokens/s with Deepseek-v4-Flash on a MacBook M5 Max. It's quite capable for coding assistant usage.

In general LLMs are bottlenecked by memory bandwidth rather than raw compute power.

But it's quantized right? It's not the same almost-free DeepSeek you get from the API.

And once the context gets large, it slows down.

"up to" 150, is doing a lot of work there.

Yes, it's quantized (4 bit). Sure, it's... not quite as good as what's on offer via API. And sure, "up to" does a lot of work (I don't have an average/median for you but it feels fast to me).

But it's usable, fully local, fully private, and has no subscriptions and no operating costs other than electricity.

I mean it’d take minutes of research to realize people are successfully and efficiently running 4-bit quantized GLM 5.2 on MacStudio 512GB M3 Ultras at over 60 tok/s. K2 2.7 is quite literally designed for 4 bit quantization and runs even better.

This is already a thing

Reply on news.ycombinator.com