nvidia/Llama-3.3-Nemotron-70B-Select

The nvidia/Llama-3.3-Nemotron-70B-Select is a 70 billion parameter large language model developed by NVIDIA, built upon the Meta-Llama-3.3-70B-Instruct foundation. It is specifically fine-tuned using scaled Bradley-Terry modeling to select the most helpful LLM-generated responses to user queries. This model is designed to improve performance in general-domain, open-ended tasks by identifying high-quality outputs, making it suitable for integration into Inference-Time-Scaling systems.