Autonomous networks—intelligent, self-managing telecom operating systems—are shifting from a future vision to a top strategic priority for current telecom operators. In NVIDIA’s newly released research report, Current Status and Trends of AI in the Telecommunications Industry, network automation has emerged as the AI application scenario with the highest return on investment and greatest emphasis.

It is important to emphasize that automation and autonomy are distinct concepts. Autonomous networks can not only execute predefined workflows but also understand operator intent, weigh various factors, and autonomously decide on the appropriate actions. The key to enabling this transformation lies in inference models and AI agents fine-tuned on telecom data.

To achieve true network autonomy, an end-to-end intelligent system must be built, with key components including telecom network models, interoperable AI agents, and network simulation tools for validating operational outcomes.

On the eve of Mobile World Congress Barcelona (MWC 2026), NVIDIA announced an open large-scale telecom model (LTM) based on NVIDIA Nemotron, a comprehensive guide for building network operations reasoning agents, and a new NVIDIA Blueprint for energy efficiency and network configuration. This blueprint helps operators progress toward autonomous operations through multi-agent orchestration.

Furthermore, as part of GSMA’s newly launched Open Telco AI initiative, NVIDIA will release this new LTM model, implementation guide, and agentic AI blueprint in open-source form via GSMA—the authoritative body for the mobile communications industry—for use across the entire sector.

Open Nemotron 3 Large Telecom Model Brings Reasoning Capabilities to the Telecommunications Sector

To successfully deploy generative AI and agentic AI at scale in telecom operations, AI models must understand the specialized language of the telecommunications domain and perform logical reasoning over complex workflows. To this end, NVIDIA has collaborated with AdaptKey AI to launch a new open-source NVIDIA Nemotron LTM with 30 billion parameters, enabling telecom operators worldwide to build autonomous networks.

Built on the NVIDIA Nemotron 3 series of foundational models, this model has been fine-tuned by AdaptKey AI using open telecom datasets including industry-standard documents and synthetic logs. It is specifically optimized to understand telecom terminology and reason through workflows such as fault isolation, remediation planning, and change validation.

As an open model, Nemotron LTM provides telecom operators with full transparency into the model’s training methodology and data sources. This allows for secure, rapid deployment in local network environments, and the direct construction and operation of AI agents. At the same time, operators can safely integrate their own network and operational data to adjust and extend the model’s telecom reasoning capabilities, advancing steadily toward autonomous operations without sacrificing data control or security.