StableVicuna is one of the first large-scale open-source chatbots to utilize Reinforcement Learning from Human Feedback (RLHF) for improved alignment.