Subscribe
Sign in
Vectara: RAG as a Service
p/vectaraThe Trusted GenAI Platform for All Builders
Start new thread
Shane Connelly

Vectara API v2 - REST API simplifies Retrieval Augmented Generation

by•

Vectara's API v2 brings an incredibly simple and easy to use retrieval augmented generation API to the market. Forget about trying to figure out text extraction, segmentation, embedding models, and hybrid search: Vectara handles it all for you

Add a comment

Replies

Best
Shane Connelly
Maker
📌
Hey all! We heard the feedback from users that existing GenAI/RAG (Retrieval Augmented Generation) tools were too hard to use. That's why we launched v2 of our API. In API v2, you can craft a RAG request and integrate it into your application with a single HTTP call in less than 10 lines of code. You can add whatever enterprise data you have, whether it's in a rich file format like a PDF or word document or a developer-friendly format like JSON or markdown. Vectara simplifies detecting and reducing hallucinations by integrating its own adversarial LLM (the Hughes Hallucination Evaluation Model, created by Vectara) that can validate the accuracy of summarization tasks. Vectara works in dozens of languages and can produce RAG results in milliseconds to single-digit seconds. We'd love to see how you'd use an end-to-end RAG platform!