---
title: "RAG Patterns"
description: "The retrieval pipeline from chunking through reranking to agentic and graph-augmented RAG - condensed current practice"
url: "https://agentsurface.dev/docs/agent-retrieval/rag-patterns"
index: https://agentsurface.dev/llms.txt
lastVerified: 2026-10-29
lastModified: 2026-10-29T11:29:43.000Z
---

RAG comes in many shapes. Treat each named variant below as a design move; most production systems combine several: hybrid retrieval, reranking, contextual chunks, and evaluation.

## Production default pipeline [#production-default-pipeline]

1. Ingest with stable document IDs, metadata, permissions, and source hashes.
2. Chunk semantically (512–1,024 tokens, 10–30% overlap is a reasonable starting default); add contextual chunk descriptions where ambiguity is common.
3. Retrieve with hybrid search: dense embeddings plus a sparse (BM25 or full-text) index, fused with Reciprocal Rank Fusion or a database-native hybrid query.
4. Rerank the top 50–100 candidates down to 5–10. Rerankers are separate models; Cohere Rerank 5 (released 2026-10-21) is one dated example, with `rerank-v5.0-pro` at $0.05 per 1M tokens and `rerank-v5.0-fast` at $0.02, per the announcement. It is an option, not a required component.
5. Enforce metadata and authorization filters during retrieval.
6. Generate with citations back to source chunks.
7. Evaluate recall, ranking quality, faithfulness, latency, and cost - see [Retrieval Evals](https://agentsurface.dev/docs/agent-retrieval/retrieval-evals).

Keep chunk metadata (`document_id`, `source`, `section`, `tenant_id`, `updated_at`, `content_hash`) with every indexed unit - chunking without provenance produces unattributable, unfilterable results.

## Pattern taxonomy [#pattern-taxonomy]

| Pattern                                                                                                    | Use when                                                          | Avoid when                                                                                                                                           |
| ---------------------------------------------------------------------------------------------------------- | ----------------------------------------------------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------- |
| **Naive** (embed, retrieve top-k, generate)                                                                | Prototypes, small clean corpora                                   | Production - fails on exact identifiers and multi-hop questions                                                                                      |
| **Hybrid** (BM25 + dense + fusion)                                                                         | Corpus has product names, codes, or IDs alongside prose           | Rarely - this is the default production stance                                                                                                       |
| **Reranked**                                                                                               | Retrieved context is noisy                                        | Ingestion or chunking is broken - reranking cannot recover a missing passage                                                                         |
| **Contextual** (Anthropic's Contextual Retrieval pattern: prepend a chunk-context summary before indexing) | Chunks are ambiguous without surrounding document context         | Index-time LLM cost or staleness of generated context is unacceptable                                                                                |
| **Parent-child**                                                                                           | Small chunks rank well but generation needs a full section        | -                                                                                                                                                    |
| **Hierarchical / RAPTOR**                                                                                  | Long documents need both local facts and document-level synthesis | Documents change frequently, or exact-source traceability matters more than abstraction                                                              |
| **HyDE / multi-query / RAG-Fusion**                                                                        | Queries are short, vague, or phrased differently from source docs | Latency and model cost outweigh the gain on simple lookups                                                                                           |
| **GraphRAG**                                                                                               | Relationships are first-class, multi-hop, and worth maintaining   | Entity resolution is unreliable, or hybrid retrieval already answers most questions - see [Knowledge Graphs](https://agentsurface.dev/docs/agent-retrieval/knowledge-graphs) |
| **Agentic RAG** (agent plans subqueries, chooses tools, iterates)                                          | Multi-step questions spanning documents, tables, and APIs         | Simple FAQ-style lookups - adds latency, cost, and failure surface                                                                                   |
| **Self-RAG / Corrective RAG**                                                                              | Factuality matters and retrieval quality varies by query          | You cannot afford extra model calls per query                                                                                                        |
| **Multimodal RAG** (ColPali-style late interaction)                                                        | Source truth is in charts, tables, scans, or slides               | OCR-based chunking already answers most queries                                                                                                      |
| **Structured/tool-backed** (SQL, OpenAPI, MCP tools)                                                       | The answer depends on live state, joins, or permissions           | Never; the retrieval layer here is a live query against a system of record                                                                           |

## Hybrid retrieval and fusion [#hybrid-retrieval-and-fusion]

```typescript
function reciprocalRankFusion(resultLists: SearchResult[][], limit = 50, k = 60) {
  const scores = new Map<string, { score: number; result: SearchResult }>();
  for (const results of resultLists) {
    results.forEach((result, index) => {
      const existing = scores.get(result.id);
      scores.set(result.id, {
        score: (existing?.score ?? 0) + 1 / (k + index + 1),
        result,
      });
    });
  }
  return [...scores.values()].sort((a, b) => b.score - a.score).slice(0, limit);
}
```

## Agentic retrieval as a bounded escalation [#agentic-retrieval-as-a-bounded-escalation]

Use agentic retrieval (query planning, multi-hop iteration, tool selection) for multi-hop questions, schema-plus-vector queries, or answer validation. Put hard bounds around it: maximum sub-queries, maximum tool calls, a token and cost budget per request, required metadata filters, and trace IDs for replay.

## Anti-patterns [#anti-patterns]

- Naming the architecture after the newest acronym instead of the failure mode it fixes.
- Adding graph retrieval before entity resolution is reliable.
- Using agentic RAG for single-hop lookups.
- Expanding queries without reranking.
- Measuring answer quality without measuring retrieval quality.

## See also [#see-also]

- [Original RAG paper](https://arxiv.org/abs/2005.11401)
- [Anthropic Contextual Retrieval](https://www.anthropic.com/research/contextual-retrieval)
- [RAPTOR paper](https://arxiv.org/abs/2401.18059)
- [Self-RAG paper](https://arxiv.org/abs/2310.11511)
- [Corrective RAG paper](https://arxiv.org/abs/2401.15884)