by farbodtavakkoli
Open source · 872k downloads · 0 likes
OTel Embedding 109M is an embedding model specifically designed for the telecommunications sector, optimized to understand and process technical data related to industry norms, specifications, and standards. It excels in information retrieval and question-answering applications, particularly in extracting relevant insights from complex documents such as 3GPP specifications or RFCs. Trained on data validated by over 200 industry experts, it ensures enhanced accuracy and relevance in professional contexts. The model stands out for its ability to improve Retrieval-Augmented Generation (RAG) systems tailored to telecommunications, delivering more reliable results for users. It seamlessly integrates with knowledge analysis and management tools in demanding technical environments.
OTel-Embedding-109M is a telecom-specialized embedding model fine-tuned on telecommunications domain data. It is part of the OTel Family of Models, an open-source initiative to build industry-standard AI models for the global telecommunications sector.
| Attribute | Value |
|---|---|
| Base Model | sentence-transformers/all-mpnet-base-v2 |
| Parameters | 109M |
| Training Method | Full parameter fine-tuning |
| Language | English |
| License | Apache 2.0 |
The model was trained on high-quality telecom-focused data curated by 200+ domain experts from organizations including AT&T, RelationalAI, AMD, GSMA, Purdue University, Khalifa University, University of Leeds, Yale University, The University of Texas at Dallas, NetoAI, and MantisNLP.
Data Sources:
This model is optimized for:
@misc{otel2026,
title={OTel: Open Telco AI Models},
author={Tavakkoli, Farbod and Diamos, Gregory and Paulk, Roderic and Terrazas, Jorden},
year={2026},
url={https://huggingface.co/farbodtavakkoli}
}
If you have any technical questions, please feel free to reach out to [email protected] or [email protected]