Telecom RAG fine-tune
According to the model card, this is a context-grounded telecom model full-parameter fine-tuned on OTel telecommunications data.
Open Source Model Profile · farbodtavakkoli
OTel-LLM-3B-IT
OTel-LLM-3B-IT is a 4.25B-parameter Mistral3 telecom RAG fine-tune from farbodtavakkoli. Its card documents OTel-domain tuning under Apache-2.0 licensing.
- Publisher
- farbodtavakkoli
- Task
- text-generation
- Model type
- mistral3
- License
- apache-2.0
- Library
- Unknown
- Publication status
- Accepted · not indexed
Model overview
OTel-LLM-3B-IT is published by farbodtavakkoli as a Mistral3 text-generation model for telecommunications. The captured configuration identifies Mistral3ForConditionalGeneration with model type mistral3, and Safetensors metadata reports 4,251,743,232 parameters. According to the model card, it is a context-grounded telecom model full-parameter fine-tuned on OTel data from a Mistral-3-3B base.
Recorded capabilities
Mistral3 4.25B-parameter record
Captured configuration records Mistral3ForConditionalGeneration with model type mistral3 and about 4.25B parameters.
Publisher-reported correctness gains
According to the model card, OTel fine-tuning lifts context-grounded correctness, with this checkpoint reported at 64.4% versus 56.9% for the base.
Domain RAG training setup
According to the model card, training used telecom data curated by domain experts with RAG-style prompt and completion rows, a 1500-token sequence length, and BF16 precision.
Use cases in the source record
- Context-grounded telecom RAG workflows that pair retrieved passages with the publisher's documented telecom prompt format.
- Abstention-aware telecom question answering using the documented prompt policy or a dedicated Safety variant where context is missing.
Limitations and unknowns
- According to the model card, answer quality depends on the retriever, reranker, context, and prompt policy, and the model is not optimized for unrestricted context-free question answering.
- No context-window value was extracted from this record beyond the 1500-token training sequence length.
- Reported evaluation gains are publisher claims judged with GPT-4o mini and were not independently verified by Ethen.
- Provider state is historical snapshot data and should be refreshed before being presented as current.
Source and provenance
Source: farbodtavakkoli/OTel-LLM-3B-IT
Captured: Unknown. Processed: 2026-09-07T19:35:56.062441+00:00.
OTel-LLM-3B-IT OTel-LLM-3B-IT is a context-grounded telecom language model full-parameter fine-tuned on OTel telecommunications data. It is part of the OTel Family of Models , an open-source initiative to build reference AI resources for the global telecommunications sector. Across the core OTel LLM baselines, OTel fine-tuning improves context-grounded correctness over the base checkpoints by +3.7 to +10.0 percentage points. Community Use As of June 23, 2026, the released OTel models had more than 18 million downloads, and the Open Telco AI project had received 157+ pieces of media coverage worldwide. Model Details Attribute Value B…
F001F002F003F004F005F006F007F008F009F010F011F013F014F015F017F018F019F021