IsmaelMousa/Qwen2.5-3B-Instruct-EngSaf-628K
TEXT GENERATIONConcurrency Cost:1Model Size:3.1BQuant:BF16Ctx Length:32kPublished:May 24, 2025License:apache-2.0Architecture:Transformer Open Weights Cold

IsmaelMousa/Qwen2.5-3B-Instruct-EngSaf-628K is a 3.1 billion parameter instruction-tuned causal language model, fine-tuned by Ismael Mousa from the Qwen2.5-3B-Instruct architecture. This model is specifically optimized for automatic essay grading, leveraging the EngSAF-628K dataset. It excels at evaluating student answers and generating rationales based on provided mark schemes, making it suitable for educational assessment tasks.

Loading preview...