andyid/legal-chatbot-qwen-1.5b-ft

TEXT GENERATIONConcurrent Unit Cost:1Model Size:1.5BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Jun 30, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

The andyid/legal-chatbot-qwen-1.5b-ft is a 1.5 billion parameter Qwen2.5-based causal language model developed by andyid. Fine-tuned from unsloth/Qwen2.5-1.5B-Instruct-bnb-4bit, this model is optimized for legal chatbot applications. It leverages Unsloth and Huggingface's TRL library for faster training, offering a 32768 token context length.

Loading preview...

Overview

The andyid/legal-chatbot-qwen-1.5b-ft is a specialized 1.5 billion parameter language model, fine-tuned from the Qwen2.5-1.5B-Instruct-bnb-4bit base model. Developed by andyid, this model is specifically designed for applications within the legal domain, aiming to function as a legal chatbot.

Key Capabilities

  • Legal Domain Specialization: Optimized for understanding and generating text relevant to legal queries and discussions.
  • Efficient Training: Utilizes Unsloth and Huggingface's TRL library, enabling significantly faster fine-tuning (2x speed improvement).
  • Qwen2.5 Architecture: Benefits from the robust architecture of the Qwen2.5 series, providing a strong foundation for language understanding and generation.
  • Extended Context Window: Supports a substantial context length of 32768 tokens, allowing for processing longer legal documents or complex conversational histories.

Good For

  • Legal Chatbots: Ideal for developing conversational AI agents that can assist with legal information, answer legal questions, or guide users through legal processes.
  • Legal Information Retrieval: Can be adapted for tasks involving extracting specific information from legal texts.
  • Legal Document Analysis: Potentially useful for preliminary analysis or summarization of legal documents within its context window limits.

This model is released under the Apache-2.0 license, making it suitable for a wide range of applications.