alifkha7/qwen2.5-1.5b-legal-id-qlora

TEXT GENERATIONPricing:Input $0.04 / Cached $0.008 / Output $0.08Concurrent Unit Cost:1Model Size:1.5BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Oct 16, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

The alifkha7/qwen2.5-1.5b-legal-id-qlora is a 1.5 billion parameter language model, fine-tuned by alifkha7 using QLoRA on the Qwen2.5-1.5B base model. It is specifically optimized for Indonesian legal consultation and question-answering tasks, leveraging a 32,768 token context length. This model excels in providing specialized legal information within the Indonesian context, making it suitable for applications requiring domain-specific legal intelligence.

Loading preview...

Overview

This model, alifkha7/qwen2.5-1.5b-legal-id-qlora, is a specialized large language model developed by alifkha7. It is built upon the Qwen/Qwen2.5-1.5B base model, featuring 1.5 billion parameters and a substantial 32,768 token context length. The model has been fine-tuned using the QLoRA method, which involves 4-bit quantization combined with LoRA adapters, to enhance its performance for specific applications.

Key Capabilities

  • Indonesian Legal Consultation: Designed to provide relevant information and guidance on Indonesian legal matters.
  • Legal Question Answering (QA): Capable of answering questions related to Indonesian law with improved accuracy and context.
  • Language Specificity: Optimized for the Indonesian language, ensuring better understanding and generation of legal texts in this context.
  • Efficient Fine-tuning: Utilizes QLoRA for efficient adaptation, making it a resource-effective solution for domain-specific tasks.

Good for

  • Developers building applications that require specialized legal AI assistance for Indonesian users.
  • Creating chatbots or virtual assistants focused on Indonesian legal queries.
  • Research and development in the field of AI for legal tech within Indonesia.

Finetunes
1 models