Vectara, a retrieval augmented generation (RAG) company, has introduced Vectara Portal, an open-source environment allowing users to build AI applications to interact with their data.

Vectara Portal enables users to create custom applications for semantic search, summarization, or conversational chat assistance without writing code. Users can upload documents to ground the AI in their data, and the tool uses Vectara's RAG-as-a-service platform to index and process the information. The tool is a hosted app and an open-source offering under the Apache 2.0 license.

Vectara aims to increase sign-ups and generate interest in its main RAG-as-a-service offering, which could potentially lead to more enterprise customers.

Analyst QuickTake: Vectara Portal’s launch builds on the company’s USD 25 million Series A funding round in July 2024 . Funds were earmarked to bolster its product, ramp up go-to-market resources, and expand offerings in Australia and the EMEA regions. The company also broadened its product use cases by introducing a new Mockingbird LLM engineered to reduce hallucinations and improve structured output in RAG applications.