nvidia/DLER-R1-1.5B-Research

The nvidia/DLER-R1-1.5B-Research model is an ultra-efficient 1.5 billion parameter reasoning model developed by NVIDIA. It is specifically designed for challenging tasks such as mathematics, programming, and scientific problem-solving. Trained with the DLER algorithm on the agentica-org/DeepScaleR-Preview-Dataset, this model achieves significant efficiency gains, reducing average response length by nearly 80% across diverse mathematical benchmarks while improving accuracy compared to similar models.