A Singapore-based startup raised $350 million to build hardware for agentic AI.
Singapore-based company Acrab has closed a funding round of over $350 million. The funds are aimed at developing specialized computing infrastructure for artificial intelligence agent systems. This is one of the largest investments in the AI infrastructure segment in the Asia-Pacific region in recent quarters.
Founded in 2024, Acrab positions itself as a developer of a full-cycle architecture. The company's portfolio includes its own AI chips, solutions for local deployment of large language models (LLMs), operating systems, multimodal interfaces, and agent management technologies. The startup's first product was the GΞLIX platform, designed to run LLMs on local hardware for agent AI tasks.
According to the team, GΞLIX has already been validated in real-world conditions and is at the stage of preparation for its first industrial deployment and mass production. The funding received will be directed towards further platform development, deepening research in computing systems, expanding the partner network, and entering international markets.
My comment: The fact that such significant capital is being raised by a startup less than a year old indicates a pressing market need for specialized hardware for agent AI. Major players are still focused on cloud solutions, while local agent deployment remains a bottleneck. If Acrab can truly scale GΞLIX to industrial volumes, it could shift the balance of power in the decentralized AI computing segment.