DayNews.ai

DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning

Open-source reasoning model claimed to match o1 performance at 1/20th training cost