kevinadityaikhsan/llama-3.2-3b-legal-id-grpo

TEXT GENERATIONConcurrent Unit Cost:1Model Size:3.2BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Jul 13, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

The kevinadityaikhsan/llama-3.2-3b-legal-id-grpo is a 3.2 billion parameter Llama-based model developed by kevinadityaikhsan. This model is a finetuned version of kevinadityaikhsan/llama-3.2-3b-legal-id-sft, specifically optimized for legal tasks in Indonesian. It was trained using Unsloth and Huggingface's TRL library, achieving 2x faster training speeds.

Loading preview...

Model Overview

The kevinadityaikhsan/llama-3.2-3b-legal-id-grpo is a 3.2 billion parameter Llama-based language model developed by kevinadityaikhsan. It is a finetuned iteration of the kevinadityaikhsan/llama-3.2-3b-legal-id-sft model, specifically designed for legal applications within the Indonesian context.

Key Characteristics

  • Architecture: Based on the Llama model family.
  • Parameter Count: 3.2 billion parameters, offering a balance between performance and computational efficiency.
  • Training Efficiency: The model was finetuned using Unsloth and Huggingface's TRL library, resulting in a 2x acceleration in the training process.
  • Specialization: This model is specialized for legal tasks in Indonesian, indicating its potential for applications requiring domain-specific understanding in this language.

Use Cases

This model is particularly well-suited for:

  • Indonesian Legal Text Analysis: Tasks such as legal document summarization, clause extraction, or legal question answering in Indonesian.
  • Legal Research Assistance: Aiding legal professionals in navigating and understanding Indonesian legal texts.
  • Domain-Specific Applications: Any application requiring a nuanced understanding of Indonesian legal terminology and concepts.