Search ⌘K Start a project →

Foundations

Fine-Tuning (vs RAG)

Fine-tuning trains a model's weights on your own examples, changing its behavior directly.

That is different from RAG, which supplies facts at query time instead of changing the model.

The change might be tone, format, or a narrow skill; fine-tuning teaches how to respond, RAG supplies what to respond with.

Using fine-tuning to fix a knowledge or freshness problem that RAG would solve cheaper is a common build-vs-buy mistake.

Insight: Build, buy, or skip

← Back to the full glossary

Related terms

Eval Suite (Evals) An eval suite is a versioned set of test cases that measures whether outputs are good. Inference Cost Inference cost is what it costs to run a trained model on a request.

More in Foundations

Retrieval-Augmented Generation (RAG) Retrieval-Augmented Generation (RAG) retrieves relevant passages at query time and feeds them into an LLM's context. Embeddings Embeddings are numeric vectors that place similar content close together in vector space. Context Window The context window is the maximum text, measured in tokens, a model can consider at once. Token A token is the basic unit of text an LLM reads and writes.

Source: https://customlabs.io/glossary/fine-tuning/