GSMA Open-Telco LLM Benchmarks

Advancing AI for the Telecom Industry

Large Language Models (LLMs) struggle with telecom-specific knowledge, often producing inaccurate technical responses, hallucinating regulatory details, and failing at critical network troubleshooting tasks. As the telecom industry increasingly integrates AI-driven solutions, ensuring these models understand telco-specific challenges is crucial to preventing regulatory risks, poor user experiences, and wasted investments.

The GSMA Open-Telco LLM Benchmark is an open-source initiative designed to enhance LLM performance in telecommunications. Led by GSMA and supported by key industry partners the benchmark evaluates AI models against real-world telecom use cases. By fostering industry-wide collaboration, this initiative aims to set the standard for AI-driven telecom intelligence, ensuring that models are optimised for accuracy, efficiency, and safety.

The initiative will officially launch at MWC, with ongoing development driven by global telecom leaders. The Open-Telco LLM Benchmarks will serve as the central hub for evaluating and improving AI models tailored for telecom applications.

Want to get involved? Share your Gen AI telco use cases and requirements to aiusecase@gsma.com and help shape the future of AI in telecom.