People running LLMs aren’t the target. People who use things like ChatGPT and CoPilot on low power PCs who may benefit from edge inference acceleration are. Every major LLM dreams of offloading compute on the end users. It saves them tons of money.
Intel sees the AI market as the way forward. NVIDIA’s AI business eclipses its graphics business by an order of magnitude now, and Intel wants in. They know that they rule the integrated graphics market, and can leverage that position to drive growth with things like edge processing for CoPilot.
Servers have had memory mirroring as a feature for years. This seems like a cool extension of that technology. It would be an advantage in some systems where scaling out isn’t an option and single node availability needs to be as high as possible.
It probably runs a completely custom instruction set which makes it incompatible with current architectures. Current manufacturers are designing chips that are operable with popular instruction sets.
People running LLMs aren’t the target. People who use things like ChatGPT and CoPilot on low power PCs who may benefit from edge inference acceleration are. Every major LLM dreams of offloading compute on the end users. It saves them tons of money.