GemSUra-edu: HCMUT FAQ Assistant

GemSUra-edu is a 2.6 billion parameter large language model, fine-tuned by IAmSkyDra specifically for answering frequently asked questions (FAQs) related to Ho Chi Minh City University of Technology (HCMUT). It builds upon the foundational GemSUra 2B model, originally developed by the URA research group at HCMUT.

Key Capabilities

• Specialized Knowledge: Optimized to provide accurate and relevant answers to queries based on HCMUT's institutional FAQs.
• Contextual Understanding: Utilizes an 8192-token context length, enabling it to process and understand longer, more complex queries.
• Efficient Inference: Supports inference with unsloth for higher speed, as well as standard transformers library for flexibility.

Good For

• Educational Institutions: Ideal for deployment in university settings to automate responses to common student and administrative inquiries.
• Information Retrieval: Excellent for specific domain-focused question-answering tasks where the knowledge base is well-defined.
• Local Device Deployment: The model can be quantized to 8 bits for efficient deployment on local devices, making it suitable for resource-constrained environments.