Ashling Technologies has confirmed that its first AI ASIC project is entering mass production, a milestone that signals the company's transition from development to commercial deployment. The vendor expects this ramp-up to drive significant revenue growth in the near term.
Chairman confirms first AI ASIC project entering mass production phase with significant revenue growth expected this year.
The company's chairman, Shen Qiangming, stated during a May 26 earnings call that while ASICs still lag behind GPUs in total market size, their current growth momentum is stronger. He projected that the compound annual growth rate (CAGR) for ASICs will eventually surpass that of general-purpose graphics processors.
Ashling's North American client is preparing to ship its 3nm AI accelerator at scale starting in Q3 2024. The firm also aims to complete tape-out for a 2nm project by the end of this year, positioning itself ahead of competitors in advanced node development.
MediaTek has adjusted its financial outlook upward, confirming that its inaugural ASIC initiative will contribute $2.0 billion USD in AI revenue for the current year. This figure represents a direct doubling from the company's previous forecast of $1.0 billion, reflecting confidence in early production yields and customer adoption.
Creative Electronics notes that ASICs deliver significantly better performance per watt than standard GPUs when handling inference workloads. As AI computing shifts toward edge devices and end-user applications, this efficiency advantage could drive shipment volumes that eventually overtake the broader GPU market.
Amazon's Trainium 3 custom chip is scheduled to begin its production ramp-up in Q2 2026, according to industry tracking data. This timeline aligns with Ashling's roadmap for advanced node deliveries and suggests sustained demand from hyperscalers for specialized AI silicon.
Discussion
0 comments
Log in to join the thread with a thoughtful take, question, or correction.