---
license: apache-2.0
datasets:
- jayasuryajsk/spoken_telugu
language:
- te
pipeline_tag: text-generation


---
# Llama-3-8b-Telugu_Romanized

Llama-3 8B finetune on synthetic data of Telugu Language.

## Model Details

### Model Description

The Llama-3-8b-Telugu-Romanized model is a language model designed for various natural language processing tasks in the Telugu language, using the Romanized script with language-mixing(English and Telugu words). It can be applied to tasks such as language generation, text completion, question answering, and other downstream NLP applications specific to the Telugu language, particularly in spoken or informal contexts where Language-mixing is prevalent.




- **Developed by:** Jaya Surya Kommireddy
- **Language(s) (NLP):** Telugu (Romanized) with code-mixing (English nouns, Telugu verbs, and mixed adjectives) and English
- **Finetuned from model:** unsloth/llama-3-8b-bnb-4bit



## Uses

The model can be used directly for tasks such as language generation, text completion, and question answering in the Telugu language with code-mixing.