RAG

RAG — explained in more detail

RAG stands for Retrieval-Augmented Generation and describes a technique that connects a language model to an external knowledge source without retraining it. The typical flow: the question is converted into a vector, a vector database returns the most similar text passages from a knowledge base, and the model receives those passages together with the original question as context. Answers therefore rest on concrete evidence rather than on the knowledge baked into the model during training alone.

Example / Practical context

A classic application is chatbots over your own documents — product manuals, internal wikis, legal texts. The model has not seen those documents during training but, thanks to RAG, can cite relevant excerpts and surface their source. This also makes it possible to bring in content created after the knowledge cutoff or internal company material without modifying the model itself.

Distinction from related concepts

Unlike fine-tuning, RAG does not change the model — the knowledge source is swappable and can be updated at any time. Compared to long context windows, where entire documents are pasted directly into the prompt, RAG scales better to large corpora because only the relevant passages are loaded per request.

Entdecke mehr

Blog

AI Workflows by Keyword: How We Make Recurring Routines Enforceable

A typed keyword triggers a fixed AI routine — and every single step must be committed before the next one appears. Why that's the actual trick.

Glossar

Reasoning Effort

Reasoning Effort is a control parameter on modern reasoning models that sets how much internal step-by-step thinking (thinking tokens) a model spends before answering — higher levels raise quality but also latency and cost.

Lexikon

LLM Hallucinations — Causes and Remedies

Why LLMs confidently invent falsehoods, what types of hallucinations exist, and which remedies actually help — RAG, source enforcement, verification.

Themenuebersicht

KI-KonzepteLLM-Grundlagen