SoftBank Corp. announced a breakthrough with its new Large Telecom Model (LTM), designed specifically for telecom operators seeking enhanced efficiency and operational agility through AI-driven automation. Built on extensive network datasets and SoftBank’s decades-long experience, the LTM acts as a generative AI foundation for network design, configuration, and management, providing operators with unprecedented speed and accuracy.
By leveraging the LTM, SoftBank successfully demonstrated specialized AI models capable of automatically generating optimal cellular base station configurations. In initial trials within Tokyo—a densely populated and high-traffic region—the fine-tuned AI models achieved over 90% accuracy in predicting configurations. Crucially, these AI-driven adjustments cut configuration times from several days down to mere minutes, drastically reducing human error and operational costs.
SoftBank’s innovative approach integrates advanced AI inference capabilities with natural language processing, allowing network operators to quickly adapt their infrastructure in response to real-time demands. Typical use cases include rapidly deploying new base stations in crowded urban environments and swiftly adjusting network configurations during high-traffic events, ensuring consistent connectivity. 
The LTM was developed in collaboration with NVIDIA, utilizing the DGX SuperPOD infrastructure for distributed training. Additionally, NVIDIA’s microservices optimization technology enhanced inference performance significantly—achieving a fivefold improvement in key metrics such as Time to First Token (TTFT) and Tokens Per Second (TPS). Further collaboration with NVIDIA on digital twin simulations promises to validate future LTM-driven network updates before deployment.
“SoftBank’s AI platform model, the ‘Large Telecom Model’ (LTM), developed for telecommunications operators, significantly transforms the processes of designing, constructing, and operating communication networks. By fine-tuning LTM, it’s possible to build AI models specialized for various processes and deploy them as agents. This not only optimizes and automates operational tasks but also enhances network performance through the tuning of wireless devices.”
Ryuji Wakikawa, Vice President and Head of SoftBank’s Research Institute of Advanced Technology (RIAT)
“Large Telecom Models are the foundation for simplifying and speeding up network operations by enabling the creation of network AI agents for specialized tasks such as network planning, network configuration and network optimization. SoftBank’s rapid innovation in developing its new LTM, leveraging NVIDIA AI technologies, sets a powerful example for telecom operators globally to redefine their network operations processes with AI.”
Chris Penrose, Vice President of Telecoms at NVIDIA
Why It Matters:
The introduction of SoftBank’s LTM reflects a pivotal shift in telecom network management, marking a step toward fully AI-integrated telecom networks—critical for 6G infrastructure. By embedding generative AI directly into operational processes, telecom providers can ensure greater reliability, responsiveness, and flexibility. The development highlights the importance of cross-sector partnerships, AI specialization, and the growing role of foundational AI models in shaping the next generation of intelligent connectivity.
