Axerela plans AI chiplets for automotive
Dutch edge AI chip designer Axelera is planning a version of its in memory computing AI architecture for automotive chiplets.
“We are already part of the imec chiplet consortium and a EU consortium to standardise RISC-V in AI acceleration,” said Fabrizio del Maffeo, CEO and co-funder of Axelera AI in the Netherlands tells eeNews Europe Automotive.
Belgian research lab imec pulled together a group to develop chiplet technology for automotive that includes European car makers.
- Axelera AI plans transformer chip variants
- imec looks to automotive chiplet programme
- RISC-V AI platform for edge vision .
- Two step flow for Axelera AI chip validation
The company is starting production of its Metis in-memory computing edge AI chip on a 12nm process at TSMC and looking at how the technology can be used as a chiplet in a package for automotive AI applications. This would have to be built on an automotive process.
“The technology will go into automotive but it’s a proof of concept and we want to explore that market in a different way,” he said. “We will show some demonstrations based on Metis that shows the potential of the technology. We have a unique key component but we are not the one to deliver it, but we can play a role,” he said.
- Major Japanese push for automotive chiplets for 2028 ..
- ASIC platform targets automotive chiplets .
- Renesas moves to chiplets for X5 automotive family
Chip designers in Aisa are already targeting chiplets for automotive applications, and Renesas is planning to use chiplets for its fifth generation R-car platform with a variety of AI accelerators alongside a major Japanese consortium to commercialise the technology..
If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :
