wisesasutresna/llama-3-indonesian-legal-bot

Hugging Face
TEXT GENERATIONConcurrency Cost:1Model Size:8BQuant:FP8Ctx Length:8kPublished:May 15, 2026License:apache-2.0Architecture:Transformer Open Weights Warm

The wisesasutresna/llama-3-indonesian-legal-bot is an 8 billion parameter Llama 3 instruction-tuned model, developed by wisesasutresna. It was fine-tuned using Unsloth and Huggingface's TRL library, specifically optimized for Indonesian legal domain tasks. This model is designed to provide specialized responses within the Indonesian legal context, leveraging its Llama 3 architecture and efficient training methodology.

Loading preview...

Overview

The wisesasutresna/llama-3-indonesian-legal-bot is an 8 billion parameter language model fine-tuned from unsloth/llama-3-8b-Instruct-bnb-4bit. Developed by wisesasutresna, this model is specifically adapted for the Indonesian legal domain.

Key Capabilities

  • Specialized Legal Understanding: Optimized for processing and generating text related to Indonesian legal contexts.
  • Efficient Fine-tuning: Leverages Unsloth and Huggingface's TRL library for faster training, indicating potential for efficient deployment and iteration.
  • Llama 3 Architecture: Built upon the robust Llama 3 instruction-tuned base model, providing a strong foundation for language understanding and generation.

Good For

  • Indonesian Legal Applications: Ideal for use cases requiring specific knowledge or generation within the Indonesian legal framework.
  • Research and Development: Suitable for developers and researchers exploring domain-specific LLM applications in legal tech.
  • Instruction-Following Tasks: Benefits from its instruction-tuned base, making it capable of following specific prompts and generating relevant legal information.