Mangara01/legal-chatbot-sft-Mangara_Haposan_Immanuel_Siagian-exp1_lr2e5_r16
TEXT GENERATIONConcurrency Cost:1Model Size:0.5BQuant:BF16Ctx Length:32kPublished:May 9, 2026License:apache-2.0Architecture:Transformer Open Weights Cold
The Mangara01/legal-chatbot-sft-Mangara_Haposan_Immanuel_Siagian-exp1_lr2e5_r16 is a 0.5 billion parameter Qwen2.5-Instruct model developed by Mangara01. Fine-tuned from unsloth/Qwen2.5-0.5B-Instruct-bnb-4bit, this model was trained using Unsloth and Huggingface's TRL library, achieving 2x faster training. It is specifically optimized for legal chatbot applications, leveraging its efficient training methodology and compact size for specialized tasks.
Loading preview...
Model Overview
This model, developed by Mangara01, is a 0.5 billion parameter Qwen2.5-Instruct variant, specifically fine-tuned for legal chatbot applications. It was built upon the unsloth/Qwen2.5-0.5B-Instruct-bnb-4bit base model.
Key Characteristics
- Efficient Training: The model was trained using Unsloth and Huggingface's TRL library, resulting in a 2x faster training process compared to standard methods.
- Compact Size: With 0.5 billion parameters, it offers a balance between performance and computational efficiency, making it suitable for deployment in resource-constrained environments.
- Specialized Fine-tuning: Its fine-tuning is geared towards legal chatbot functionalities, suggesting enhanced performance in understanding and generating legal-domain specific text.
Use Cases
This model is particularly well-suited for:
- Developing legal chatbots that require domain-specific understanding.
- Applications where fast training and inference are critical due to its Unsloth-optimized development.
- Scenarios requiring a smaller, efficient language model for legal text processing.