AI Accelerator Chiplets
Filter
2
Chiplet
(1
-
2)
-
Machine learning accelerator chiplet implemented in 12nm
- 3 TOPS peak performance AI accelerator
- Hardware accelerated ReLU/Softmax nonlinearities
- Hardware accelerated transpose
-
Hub Chiplet
- Rich pluggable SoC infrastructure that can help accelerators and CXL solutions reach the market 5x-10x faster with over 10x cost reduction