Qwen/Qwen2.5-Math-72B-Instruct
Qwen/Qwen2.5-Math-72B-Instruct is a 72.7 billion parameter instruction-tuned causal language model developed by Qwen, specifically optimized for solving mathematical problems. This model excels in both English and Chinese mathematics, supporting Chain-of-Thought (CoT) and Tool-integrated Reasoning (TIR) for enhanced computational accuracy and symbolic manipulation. It is designed to be a mathematical expert model, building upon the Qwen2-Math series with significant performance improvements on mathematical benchmarks.
No reviews yet. Be the first to review!