farbodtavakkoli/OTel-LLM-12B-IT

OTel-LLM-12B-IT: A Specialized Telecom Language Model

OTel-LLM-12B-IT, developed by farbodtavakkoli, is a 12 billion parameter language model built upon google/gemma-3-12b-it. It has undergone full-parameter fine-tuning using the OTel-LLM dataset, which comprises telecommunications-specific data curated by over 100 domain experts. This specialization makes it a key component of the OTel Family of Models, an initiative focused on open-source AI resources for the global telecom sector.

Key Capabilities & Differentiators

• Telecom Domain Expertise: Specifically trained on a high-quality, filtered dataset of 326,767 telecom-focused examples, including arXiv papers, 3GPP standards, GSMA documents, and O-RAN specifications.
• Enhanced Context-Grounded Correctness: Demonstrates a notable improvement of +5.0 percentage points in LLM-as-judge correctness for context-grounded answers compared to its base model.
• RAG Optimization: Primarily intended for use in Retrieval-Augmented Generation (RAG) pipelines, where it generates answers grounded in provided telecom context.
• Apache 2.0 License: Released under an open-source license, promoting community use and development.

Intended Use Cases

• Context-Grounded Telecom QA: Ideal for answering questions within the telecommunications domain when provided with relevant retrieved context.
• RAG System Integration: Designed to be integrated into RAG systems for specialized telecom information retrieval and synthesis.

It's important to note that this model is domain-specific and English-only, not optimized for unrestricted, context-free general question answering. Users should verify generated content, especially for critical operational or regulatory applications.