OTel-LLM-8.2B-IT: A Telecom-Specialized Language Model

OTel-LLM-8.2B-IT is an 8.2 billion parameter language model, fine-tuned from the Qwen3-8B base model, specifically designed for the telecommunications industry. It is part of the OTel Family of Models, an open-source initiative by farbodtavakkoli aimed at developing industry-standard AI for global telecom.

Key Capabilities

• Domain Expertise: Specialized in telecommunications, trained on extensive data including GSMA Permanent Reference Documents, 3GPP Specifications, O-RAN Documentation, and RFC Series.
• High-Quality Training Data: Curated by over 200 domain experts from leading organizations like AT&T, AMD, GSMA, and various universities.
• Instruction-Tuned: Optimized for specific tasks within the telecom domain.

Intended Use Cases

• RAG Applications: Ideal for Retrieval Augmented Generation (RAG) systems in telecommunications.
• Question Answering: Excels at answering questions related to telecom specifications, standards, and industry documentation.

Training Details

The model underwent full parameter fine-tuning using the ScalarLM framework, leveraging compute resources from TensorWave (AMD GPUs) and Azure (NVIDIA GPUs). It is licensed under Apache 2.0.