Skip to main content
Use these tutorials to match the Luna Studio SDK workflow to your dataset shape and metric type.

Pick the tutorial that matches your inputs

LLM spans without RAG

Build a metric for standard LLM spans without retrieved context.

LLM spans with RAG

Train against span inputs that include retrieved context.

LLM spans with tools

Cover agentic traces that invoke tools during execution.

Retriever spans

Work with retrieval-only span inputs.

Trace input / output only

Start from simple trace-level inputs and outputs.

Using a preset metric

Adapt a Galileo preset metric to your own data.

Full traces

Cover complete traces with richer structure.

List of trace inputs / outputs only

Train on lists of trace input/output pairs.

Full sessions

Work with session-level inputs spanning multiple turns.