AMD acquires Taalas to boost inference performance by etching models in silicon

I've had an endgame idea in mind for a while.

Models, probably first open weight ones like Kimi K3 class, are etched into silicon like this and sold as cartridges almost like old school game cartridges.

You buy a USB-C dongle that the cartridge goes into, or for data centers you have PCI cards that take these in slots.

Yeah, so https://chatjimmy.ai/ ... the model is crap, but the speed is amazing. Worth checking out.

Imagine the size of chip needed to 'etch' something like Qwen 3.6 27B in size.

> At 20 billion parameters per chip, you’d need just 50 accelerators to support a trillion-parameter model

I don't see any evidence that this is possible. From my understanding, the whole model needs to be on a single chip. Which rules out any popular frontier models with several trillions of parameters. Even smaller sub-frontier models have hundreds of millions of parameters, so these would be ruled out as well.

Taalas is just a phenomenal startup from Toronto. My dearest congratulations to the founders.

Edit: Lol, downvotes? Stay jelly, meanwhile Talas goes brrr.

[flagged]

[flagged]