LeeChanRX/LeeChan-LegalRights

Hugging Face
TEXT GENERATIONConcurrency Cost:1Model Size:3.1BQuant:BF16Ctx Length:32kPublished:May 15, 2026License:apache-2.0Architecture:Transformer Open Weights Warm

LeeChanRX/LeeChan-LegalRights is a 3.1 billion parameter legal language model developed by LeeChanRX, fine-tuned using supervised fine-tuning (SFT) on curated legal datasets including LegalFlow-v1, LegalBench, LEDGAR, and EURLEX. This model specializes in legal question answering, contract explanation, and basic legal reasoning, providing plain-language legal responses. With a context length of 32768 tokens, it is optimized for legal domain applications.

Loading preview...

LeeChanRX-LegalRights Overview

LeeChanRX/LeeChan-LegalRights is a 3.1 billion parameter legal language model developed by LeeChanRX, specifically designed for legal domain applications. It was fine-tuned using supervised fine-tuning (SFT) with LoRA on a comprehensive set of legal datasets, including LegalFlow-v1, LegalBench, LEDGAR, and EURLEX. The fine-tuning process involved approximately 29 million trainable parameters, representing 0.96% of the total model parameters.

Key Capabilities

  • Legal Question Answering: Provides direct answers to legal queries.
  • Contract Explanation: Simplifies complex contract clauses into understandable language.
  • Basic Legal Reasoning: Offers fundamental legal insights and analysis.
  • Plain-Language Responses: Generates legal information in an accessible format.

Good For

  • Educational Purposes: Ideal for learning and understanding legal concepts.
  • Initial Legal Information Retrieval: Useful for obtaining preliminary legal explanations.
  • Developers building legal AI tools: Provides a specialized base model for legal text processing.