goasty/Qwen3-4B-Indian-Law
goasty/Qwen3-4B-Indian-Law is a 4 billion parameter language model fine-tuned from Qwen3-4B, specifically adapted for Indian legal texts. It excels in Indian legal question answering, statutory interpretation, and case law understanding, leveraging a comprehensive corpus of Indian legal documents including judgments, constitutional law, and criminal law. This model is optimized for legal research assistance and educational applications within the Indian legal domain.
Loading preview...
Qwen3-4B Indian Law: Specialized Legal Assistant
This model, goasty/Qwen3-4B-Indian-Law, is a 4 billion parameter language model fine-tuned from the Qwen3-4B base model. Its primary distinction lies in its specialized adaptation for the Indian legal domain, achieved through LoRA + QLoRA fine-tuning on an extensive corpus of Indian legal texts.
Key Capabilities
- Indian Legal Question Answering: Provides answers to queries based on Indian legal provisions and case law.
- Statutory and Constitutional Interpretation: Assists in understanding and interpreting Indian statutes and constitutional articles.
- Case Law Understanding: Processes and summarizes Indian court judgments, including Supreme Court and High Court rulings.
- Legal Reasoning: Designed to aid in complex legal reasoning tasks across various branches of Indian law.
- Domain-Specific Knowledge: Covers areas like criminal law, evidence law, procedural law, and constitutional law, drawing from datasets such as the Indian Penal Code (IPC), Code of Criminal Procedure (CrPC), and Bharatiya Nyaya Sanhita (BNS).
Training and Data
The model was trained on a diverse and balanced dataset comprising over 300,000 samples. This includes large-scale legal instruction datasets derived from Indian court judgments, specific instruction-response pairs covering Indian legal concepts, and custom-processed QA datasets for the Constitution of India, IPC, and CrPC. A significant portion of the training data comes from 16,726 Indian Supreme Court and High Court judgment files, converted into a context-question-answer format. Dataset balancing was crucial to prevent over-representation of judgment-derived samples and ensure broad coverage across statutory and constitutional law.
Intended Use Cases
This model is suitable for:
- Legal research assistance for professionals and students.
- Educational tools for law students.
- Developing legal Q&A systems.
- Applications requiring statutory interpretation and legal document analysis within the Indian context.
It is important to note that while powerful, this model is not a substitute for licensed legal counsel and should not be relied upon for legal advice without expert review.