← AI glossary

RAG (retrieval-augmented generation)

Before answering, the model looks up relevant passages from your documents — your own knowledge, no training needed.

With RAG, a question first triggers a search through your documents (PDFs, web pages, policies); the best matching passages are then given to the model along with the question.

The model answers from your current, specific information, makes things up less often, and when your documents change you just re-index — no retraining.

Example

A Bangla assistant over every university regulation PDF that cites the source of each answer.

বাংলায়: RAG (তথ্য খুঁজে উত্তর) — উত্তর দেওয়ার আগে মডেল আপনার ডকুমেন্ট থেকে প্রাসঙ্গিক অংশ খুঁজে নেয় — ট্রেনিং ছাড়াই নিজের তথ্য জানানো।

Try it on a real AI Computer

JupyterLab opens in about a minute. Pay in taka with bKash, billed by the minute — stop whenever you like.