Private Data Now Used by LlamaIndex in Large Language Models

Former Uber research scientist, Jerry Liu, has launched a start-up called LlamaIndex, which aims to unlock the potential of large language models (LLMs) such as GPT-3 and GPT-4, by overcoming their limitations in working with private data, such as personal files. Liu and colleagues developed a framework enabling developers to integrate personal or organisational data with LLMs to facilitate question answering, summarisation, insight extraction and other functions. Liu said his project is distinct in focusing on data ingestion and management with extensive indexing options to support specific domains. The open-source project has already attracted 200,000 monthly downloads, and is set to launch an enterprise version later this year. Last month, LlamaIndex raised $8.3m in seed funding led by Greylock.