jhn9803/Qwen2.5-MATH-1.5B-Instruct-DAPO-G8
TEXT GENERATIONConcurrency Cost:1Model Size:1.5BQuant:BF16Ctx Length:32kPublished:Dec 28, 2025Architecture:Transformer Warm

The jhn9803/Qwen2.5-MATH-1.5B-Instruct-DAPO-G8 model is a 1.5 billion parameter instruction-tuned language model based on the Qwen2.5 architecture. Developed by jhn9803, it is specifically fine-tuned for mathematical reasoning tasks using the hendrycks-math-with-answers dataset. This model leverages the GRPO training method, making it particularly effective for solving complex mathematical problems.

Loading preview...