nvidia/AceReason-Nemotron-1.1-7B

NVIDIA's AceReason-Nemotron-1.1-7B is a 7.6 billion parameter language model built upon the Qwen2.5-Math-7B base, specifically optimized for advanced math and code reasoning tasks. It leverages supervised fine-tuning (SFT) followed by reinforcement learning (RL) to achieve high performance on challenging benchmarks. This model significantly outperforms its predecessor, AceReason-Nemotron-1.0-7B, setting new performance records among Qwen2.5-7B-based reasoning models.