Can I use AXYS to build Retrieval Augmented Generation (RAG) workflows for AI agents?

Yes, AXYS is purpose-built to fully support Retrieval Augmented Generation (RAG) workflows for your AI agents. The platform provides pre-built, proprietary RAG pipelines that are ready to use out of the box, enabling your agents to instantly retrieve and leverage relevant data, structured or unstructured, for more accurate and contextual responses. With AXYS’s intuitive, no-code
Read More

Can I ask free-form questions and/or predetermined questions from my data in the AXYS Chat AI interface?

Yes, the AXYS Chat AI interface lets you ask both free-form and predetermined questions from your data. For unstructured data like documents or emails, you can simply ask any free-form question—no prompt setup required. The AI understands your query and searches across all available data fields and content. For structured data, you can build prompts
Read More

What data types does AXYS support for unstructured data?

AXYS supports a wide range of unstructured data types, including PDFs, Word documents, Excel spreadsheets, plain text files, CSV files, emails, and other text-based documents. Any document containing readable text can be ingested, indexed, and analyzed by the platform, making it easy to unify and search across all your unstructured business content.

Can AXYS handle both structured and unstructured data?

Yes, AXYS is designed to handle both structured and unstructured data seamlessly. The platform can connect to databases, spreadsheets, and other structured sources, as well as process unstructured content like documents, emails, PDFs, and more. This versatility ensures you can unify, analyze, and make use of all your business data—regardless of format—within a single platform.

What is the approximate cost per question for structured and unstructured data before caching?

With AXYS, the average complex query across all your structured or unstructured data uses just 3,000 to 4,000 tokens, thanks to our highly efficient design. This translates to a typical cost of about $0.02 to $0.03 per question when using OpenAI or similar large language models. For comparison, the industry average for similar queries is
Read More