nvidia/AceMath-72B-Instruct

nvidia/AceMath-72B-Instruct is a 72.7 billion parameter instruction-tuned causal language model developed by NVIDIA, based on the Qwen2.5-Math-72B-Base architecture. It is specifically optimized for advanced mathematical reasoning tasks, excelling at solving English mathematical problems using Chain-of-Thought (CoT) reasoning. This model is designed for non-commercial use and has a context length of 131072 tokens.