yanwarpro/Qwen2.5-Legal-SFT-Dicoding-Final

TEXT GENERATIONConcurrent Unit Cost:1Model Size:1.5BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Jul 8, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

The yanwarpro/Qwen2.5-Legal-SFT-Dicoding-Final is a 1.5 billion parameter Qwen2.5 instruction-tuned causal language model developed by yanwarpro. It was fine-tuned using Unsloth and Huggingface's TRL library, indicating an optimization for efficient training. This model is specifically designed for legal-related tasks, leveraging its fine-tuning to process and generate legal text effectively. Its 32768 token context length supports handling extensive legal documents.

Loading preview...

Model Overview

The yanwarpro/Qwen2.5-Legal-SFT-Dicoding-Final is a 1.5 billion parameter instruction-tuned language model based on the Qwen2.5 architecture. Developed by yanwarpro, this model was fine-tuned from unsloth/qwen2.5-1.5b-instruct-unsloth-bnb-4bit with a focus on legal applications.

Key Capabilities

  • Legal Domain Specialization: The model has undergone specific fine-tuning for legal tasks, suggesting enhanced performance in understanding and generating legal-centric content.
  • Efficient Training: It leverages Unsloth and Huggingface's TRL library, which enabled a 2x faster training process.
  • Context Handling: With a context length of 32768 tokens, it is capable of processing and retaining information from lengthy legal documents.

Good For

  • Legal Text Analysis: Ideal for tasks requiring comprehension or generation of legal documents, contracts, or case summaries.
  • Legal Research Assistance: Can potentially aid in sifting through legal information and providing relevant insights.
  • Applications requiring efficient Qwen2.5 fine-tuning: Demonstrates the effectiveness of Unsloth for rapid model adaptation.