AMD To Acquire Chip Startup Taalas To Boost AI Inference Capabilities

Advanced Micro Devices (AMD) will acquire chip startup Taalas for an undisclosed amount, as the company strengthens its technology for the growing market of AI inference workloads.

Toronto-based Taalas develops specialised silicon designed to reduce computing and memory bottlenecks in AI inference, the process of running trained AI models to generate responses or predictions. Founded in 2023, the company raised $169 million in February to support development of chips optimised for AI models, taking its total funding to around $219 million.

Why Inference Chips Matter

Specialised inference chips have become a key focus area for semiconductor makers as AI usage shifts from training models to real-time, high-volume deployment, with companies looking to cut computing costs. Rival Nvidia had unveiled a new central processor and AI system built on technology from Groq, another inference-focused chip startup, in March.

AMD plans to integrate Taalas' technology into its accelerator roadmap and develop system-level solutions using its Instinct graphics processing units (GPUs). Vamsi Boppana, senior vice president of AMD's Artificial Intelligence Group, said in a statement, "AMD is building a full-stack AI platform that gives customers the flexibility to deploy the right compute solutions for every AI workload."

Boppana added that Taalas' technology and engineering team would strengthen AMD's AI portfolio by "delivering differentiated inference performance and efficiency."

The Taalas deal is the latest in a series of AI-focused acquisitions by AMD. In November, the company acquired MK1, an AI software startup specialising in high-speed inference. It also purchased MEXT in June to expand its AI portfolio, and added FastFlowLM to its Artificial Intelligence Group in July.