Wiki vs RAG

Comparação mature

Wiki vs RAG

Overview

Both approaches let you query a large document collection. They differ fundamentally in when synthesis happens.

Comparison

DimensionLLM WikiSemantic RAG
How knowledge is storedPre-compiled markdown pages with cross-references already builtRaw chunks in a vector database
Finding answersRead index → follow links → synthesizeEmbed query → similarity search → assemble
Query costLow — synthesis already doneHigher — re-derives on every query
InfrastructureJust markdown filesEmbedding model + vector DB + chunking pipeline
MaintenanceRun a lint passRe-embed when content changes
Scale limit~hundreds of pages (index file navigation)Millions of documents
Setup time5 minutesHours to days
Contradiction detectionBuilt in — LLM flags on ingestManual

Verdict

Under 1000 pages → LLM Wiki. The index file is sufficient for navigation, token cost is low, setup is minimal, and the pre-compiled synthesis means every query benefits from everything ever read.

Over 100K pages → RAG. The index file becomes too large to read, and embedding-based retrieval becomes more efficient than full-index scanning.

The sweet spot: run the wiki pattern for active research (where things are being added, synthesized, and connected), then export to a vector store if the collection grows beyond the index threshold.

(Source: LLM Wiki Pattern, Compounding Knowledge)