The IPU is the first processor in the world to use Wafer-on-Wafer (WoW) 3D stacking technology, taking the proven benefits of the IPU to the next level. Featuring groundbreaking advances in compute architecture and silicon implementation, communication and memory, each Bow IPU delivers up to 350 teraFLOPS of AI compute, an impressive 40% leap forward in performance and up to 16% more power efficiency compared to the previous generation IPU.A Major Leap Forward
PCIe
- PCIe Gen4 x16
- 64GB/s bidrectional bandwidth to host
Wafer-On-Wafer
- Advanced silicon 3D stacking technology
- Closely coupled power delivery die
- Higher operating frequency and enhanced overall performance
Deep Trench Capacitors
- Efficient power delivery
- Enables increase in operational performance
IPU-Exchange™
- 11TB/s all to all IPU-Exchange
- Non-blocking, any communication pattern
IPU-Tiles™
- 1472 independent IPU-Tiles each with IPU-Core™ and In-Processor-Memory™
In-Processor-Memory™
- 900MB In-Processor-Memory per IPU
- 65.4TB/s memory bandwidth per IPU
IPU-Core™
- 1472 independent IPU-Cores
- 8832 independent program threads executing in parallel
IPU-Links™
- 10x IPU-Links
- 320GB/s chip to chip bandwidth
Click the markers on the diagram for more details about the chip
Swipe to see more