nvidia/OpenReasoning-Nemotron-1.5B

OpenReasoning-Nemotron-1.5B is a 1.5 billion parameter large language model developed by NVIDIA, based on the Qwen2.5-1.5B architecture. It is specifically post-trained for advanced reasoning tasks across mathematics, code generation, and scientific problem-solving. This model excels in generating solutions for complex reasoning challenges and supports up to 64K output tokens.