The Index-1.9B series is a family of lightweight, open-source large language models developed by Bilibili. It includes various versions, such as base, pure, chat, and character, suitable for pre-training on primarily Chinese and English language data. The models have demonstrated superior performance on multiple benchmark datasets. The models support SFT and DPO alignment, as well as RAG technology for role-playing customization, making them ideal for dialogue generation and role-playing scenarios.