farbodtavakkoli/OTel-2.0-LLM-31B-IT

Hugging Face
VISIONConcurrent Unit Cost:2Model Size:31BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Jul 23, 2026License:apache-2.0Architecture:Transformer0.0K Open Weights Warm

OTel-2.0-LLM-31B-IT by farbodtavakkoli is a 31 billion parameter instruction-tuned language model, post-trained from Gemma 4 31B-IT on approximately 440 billion telecom-specific tokens. This model is specialized for telecommunications, excelling in tasks like standards interpretation, network configuration assistance, RAG over telecom documents, and telecom-specific QnA. It is designed to support telco-grade AI workflows, offering domain knowledge for applications in network operations and product development.

Loading preview...

OTel-2.0-LLM-31B-IT: A Specialized Telecom AI Model

OTel-2.0-LLM-31B-IT is a 31 billion parameter instruction model, developed by farbodtavakkoli, that has been extensively post-trained from Gemma 4 31B-IT using approximately 440 billion telecom training tokens. This model represents the first release in the OTel 2.0 family, significantly expanding upon the original OTel effort by incorporating a much larger and more diverse standards and telecom corpus.

Key Capabilities

  • Telecom Domain Expertise: Specialized in telecommunications, trained on ~15 billion raw tokens from sources like GSMA, 3GPP, ETSI, ITU, CAMARA, O-RAN, and TM Forum.
  • Enhanced Data Preparation: Features broader data mixtures compared to OTel 1.0, including:
    • Agentic Tool Calling: Improved tool-use behavior for telecom AI workflows.
    • Direct Telecom QnA: Knowledge and factual question-answering for standards, protocols, and network concepts.
    • Instruction Following: Supports telecom-specific instructions for analysis, summarization, and operational tasks.
    • RAG & Abstention: Context-grounded answer generation from technical documents and the ability to abstain when context is insufficient.
  • Continuous Improvement: OTel 2.0 models are expected to receive weekly weight updates, indicating ongoing training and refinement.

Good For

  • Retrieval-Augmented Generation (RAG) over telecom standards and technical documentation.
  • Standards Interpretation and Summarization for major telecom bodies (3GPP, ETSI, GSMA, etc.).
  • Product Development and Network Configuration Assistance in telecom environments.
  • Telecom-specific Direct QnA where domain knowledge is critical.
  • Agentic Workflows requiring specialized tool-use in telecom AI applications.

Finetunes
1 models