theoitssurabaya/Legal-Assistant-SFT

TEXT GENERATIONConcurrent Unit Cost:1Model Size:7.6BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Jul 9, 2026License:apache-2.0Architecture:Transformer0.0K Open Weights Featherless Exclusive Cold

The theoitssurabaya/Legal-Assistant-SFT is a 7.6 billion parameter Qwen2-based causal language model, fine-tuned by theoitssurabaya. Optimized for legal assistance tasks, this model leverages Unsloth and Huggingface's TRL library for efficient training. It offers a 32768-token context length, making it suitable for processing extensive legal documents and queries.

Loading preview...

Model Overview

The theoitssurabaya/Legal-Assistant-SFT is a 7.6 billion parameter language model, fine-tuned from unsloth/qwen2.5-7b-instruct-unsloth-bnb-4bit. Developed by theoitssurabaya, this model is specifically designed for legal assistance applications.

Key Capabilities

  • Legal Domain Specialization: Fine-tuned to understand and process legal-specific language and contexts.
  • Efficient Training: Utilizes Unsloth and Huggingface's TRL library, enabling faster training compared to standard methods.
  • Qwen2 Architecture: Based on the robust Qwen2 model family, providing strong foundational language understanding.
  • Extended Context Window: Supports a 32768-token context length, beneficial for analyzing lengthy legal documents or complex case details.

Good For

  • Legal Information Retrieval: Answering questions related to legal texts.
  • Document Analysis: Processing and summarizing legal documents.
  • Legal Research Support: Assisting with preliminary legal research tasks.

This model is released under the Apache-2.0 license, allowing for broad use and modification.