When to Use - Generating embeddings inside Cypher without external Python ( ) - Batch-embedding nodes/chunks during ingestion ( ) - Calling LLMs directly in Cypher for completions or GraphRAG ( ) - Extracting structured JSON maps from LLM inside Cypher ( ) - Aggregating LLM summaries over grouped rows ( ) - Stateful chat sessions in Cypher ( ) - Counting tokens or chunking text by token limit ( , ) When NOT to Use - Python-based GraphRAG pipelines (VectorCypherRetriever, HybridCypherRetriever) → - Vector index CREATE / kNN search / SEARCH clause → - GDS embeddings (FastRP, Node2Vec) → - Fullt…