
According to AMD’s announcement on Thursday, the company has reached a definitive agreement to acquire Toronto-based AI chip startup Taalas, aiming to strengthen its position in the rapidly growing data center AI market. Financial terms of the acquisition were not disclosed.
Taalas develops specialized chips designed for AI inference, the stage where trained artificial intelligence models process user queries and generate responses. The startup’s technology uses a hardware-based approach by embedding AI models directly into chips, allowing it to deliver more efficient performance compared with traditional computing solutions.
AMD said Taalas’ technology will be integrated into its future product roadmap and could be introduced as server computing components. The acquisition is part of AMD’s broader strategy to expand its AI hardware offerings and differentiate its solutions from market leader NVIDIA.
As the world’s second-largest graphics processing unit (GPU) manufacturer, AMD has built its AI strategy around GPUs, which are highly effective at processing large-scale data for AI model training. However, as the AI industry shifts from model development toward large-scale deployment and real-time services, demand is increasing for alternative chips optimized for faster and more energy-efficient AI inference.
By adding Taalas’ specialized AI chip technology, AMD aims to provide customers with more flexible computing options for next-generation data centers and AI workloads.