Extend GPU/CPU coherency to multi-GPU cores

This invention describes a special kind of graphics processor designed to run neural networks more efficiently. It uses a shared memory system and a smart scheduler that understands individual neurons within the network. This allows different parts of the processor to work together on the neural network's calculations using a common memory pool.

Why it matters: Since 2021, the scale and complexity of neural networks, particularly large language models, have dramatically increased. This invention's focus on efficient GPU cache coherency for neural networks directly addresses the growing need for faster, more powerful AI processing hardware.

Status: Not in forceListed as no longer active. The specific reason is not in the record we hold.
How hard to build: SpecializedCustom silicon, complex hardware/software co-design.

3 AI reinvention directions for this patent

See all three

AI gives you a few directions you could take this. Pick one, and we check whether your version is different enough to patent, then write the filing.

Reinvent this with AI

Where this one came from

Every morning we go through millions of abandoned patents and pick out 15 worth your time. This was one of today’s.

This one

+10