An integrated circuit that is dedicated to training and executing neural networks, the architecture of AI software. The training side (deep learning) and execution side (inference engine) require massive amounts of calculations, and AI chips perform them in parallel trillions of times every second. See
Tensor core,
neural processing unit,
deep learning,
H100 and
Cerebras AI computer.
The Versal Chip
Along with other specialized processors, the top model of this Xilinx system-on-chip (SoC) includes AI engines (see
Versal).
(Image courtesy of Xilinx.)