◆ Category · 17 assets

RAG Advanced

Browse 17 RAG Advanced modes for AI coding agents — production-grounded, cited, installable. Part of the VIBE library.

🔎 Search within RAG Advanced
mode

agentic-rag-expert-mode

Agent loops over retrieval — tool-calling search, query decomposition, iterative refinement

View → mode

chunk-strategy-expert-mode

Fixed, recursive, semantic, proposition-based, document-aware chunking deep-dive

View → mode

colbert-expert-mode

ColBERT v2 / late interaction retrieval, MaxSim, RAGatouille, indexing tradeoffs

View → mode

contextual-retrieval-expert-mode

Anthropic's Contextual Retrieval — prepend chunk-specific context before embedding

View → mode

crag-expert-mode

Retrieval evaluator + web search fallback when knowledge base is insufficient

View → mode

graphrag-expert-mode

Microsoft GraphRAG — entity/community extraction, hierarchical summarization, local vs global queries

View → mode

hybrid-search-expert-mode

BM25 + dense vectors fused with RRF, alpha tuning, and SPLADE sparse-dense

View → mode

hyde-expert-mode

Hypothetical Document Embeddings — generate a fake answer, embed that, then retrieve

View → mode

late-chunking-expert-mode

Jina's late chunking — embed long context first, then chunk the embeddings

View → mode

multimodal-rag-expert-mode

RAG over images+text — ColPali, DSE, jina-embeddings-v4, voyage-multimodal-3, PDF chunking

View → mode

parent-child-retriever-expert-mode

Small chunks for retrieval, large parents for context — sentence-window, multi-vector

View → mode

query-rewriting-expert-mode

Query expansion, decomposition, multi-query retrieval, step-back prompting

View → mode

rag-eval-expert-mode

RAGAS metrics, needle-in-haystack, RULER, golden sets, faithfulness and context precision/recall

View → mode

raptor-expert-mode

RAPTOR — recursive abstractive summarization tree for long-document RAG

View → mode

rerank-expert-mode

Cross-encoder rerankers — Cohere Rerank, Jina, BGE, Voyage, ColBERT-as-reranker

View → mode

self-rag-expert-mode

Self-RAG — model decides when to retrieve and critiques evidence with reflection tokens

View → mode

structured-rag-expert-mode

RAG over structured data — text-to-SQL with retrieval, semantic layer, table-augmented generation

View →