Aixin Yuanzhi's next-generation high-performance AI chip has completed its tape-out, supporting multi-chip cascading to achieve full-scale inference for large models on the edge.

On August 12, Interface News learned from the 2026 semi-annual earnings conference call of Aixin Yuanzhi that the company's next-generation high-performance large-scale AI chip has completed the tape-out, significantly enhancing its computing power specifications. It is also equipped with high bandwidth and supports two-core or four-core cascading, enabling full-scale large models to perform high-performance inference at the edge.