Bell and Cohere Partner to Host LLMs on Bell AI Fabric Infrastructure

Bell Canada has forged a significant infrastructure partnership with Cohere, a leading artificial intelligence firm, to host Cohere's sophisticated large language models (LLMs) on Bell AI Fabric. This collaboration leverages Bell's robust data center capabilities, specifically utilizing its facility in Merritt, British Columbia, to provide the necessary computational resources. This strategic step is to bolster Canada's AI infrastructure and offer advanced, secure AI solutions for various sectors. The agreement extends beyond just Bell and Cohere, incorporating the expertise of Buzz High Performance Computing, a subsidiary of Vancouver-based Hive Digital Technologies Ltd., and Quebec-based Hypertec. Buzz HPC is tasked with developing and deploying an AI-native cloud layer, which will be underpinned by Hypertec's specialized hardware clusters. These clusters are designed to integrate NVIDIA accelerated computing, crucial for handling the intensive demands of production-grade AI workloads and ensuring optimal performance for Cohere's LLMs. The primary goal for Cohere in this partnership is to efficiently operate its foundational AI models and deliver secure, enterprise-grade AI solutions. These solutions are specifically tailored for government and business clients, addressing their unique requirements for data privacy, security, and performance. The initiative underscores a shared commitment among the participating companies to cultivate an environment conducive to critical AI research and development within Canadian borders. This focus on domestic infrastructure is vital for maintaining data sovereignty and fostering local technological innovation. This latest agreement represents a deepening of an existing relationship between Bell and Cohere. The two companies initially partnered in July to provide comprehensive, full-stack sovereign AI solutions for Canadian enterprise and government sectors. The expansion of this collaboration highlights a growing industry trend towards establishing robust, localized infrastructure dedicated to AI development and deployment. Such infrastructure is essential not only for the high-performance computing required by modern LLMs but also for addressing concerns around data governance and national security in the rapidly evolving AI landscape. The integration of advanced hardware and cloud services aims to provide a scalable and secure platform for the next generation of AI applications.