AMD Acquires AI Chip Startup That Embeds Models in Silicon
AMD has purchased a chip startup that hardwires AI models directly into hardware, targeting faster and more efficient AI inference.
AMD has acquired a chip startup focused on embedding artificial intelligence models directly into silicon, a move that signals the semiconductor giant's push to compete more aggressively in the rapidly expanding AI hardware market. The acquisition brings AMD a technology that bakes AI capabilities into the chip itself, rather than relying on software layers to run models at runtime.
The startup's current chip is designed to run a compact version of Meta's Llama 3.1, one of the most widely used open-source large language models available today. By hardwiring the model into the silicon, the approach promises lower latency and reduced power consumption compared with traditional inference methods that load models dynamically.
Read more 53-Year-Old Lawn and Garden Company Files Chapter 11 Bankruptcy →
The company is also developing chips capable of supporting larger and more advanced AI models, suggesting AMD sees significant room to scale the technology beyond its current form. That roadmap positions the acquisition as a long-term strategic bet, not merely a near-term product addition.
For AMD, the deal deepens its challenge to Nvidia's dominance in AI accelerator hardware, a segment that has become one of the most valuable in the entire technology industry. Hardwired AI silicon could appeal to enterprises and cloud providers seeking predictable, efficient inference at scale without the overhead of general-purpose GPU clusters.
Continue reading at US Top News and Analysis.
