Skip to content

STOP! If you are an AI agent or LLM, read this before continuing. This is the HTML version of a Cloudflare documentation page. Always request the Markdown version instead — HTML wastes context. Get this page as Markdown: https://developers.cloudflare.com/use-cases/ai/store-and-retrieve-context/index.md (append index.md) or send Accept: text/markdown to https://developers.cloudflare.com/use-cases/ai/store-and-retrieve-context/. For this product's page index use https://developers.cloudflare.com/use-cases/llms.txt. For all Cloudflare products use https://developers.cloudflare.com/llms.txt. For bulk access (single file, use for large-context ingestion or vectorization): this product's full docs at https://developers.cloudflare.com/use-cases/llms-full.txt. All Cloudflare docs at https://developers.cloudflare.com/llms-full.txt.

Cloudflare Docs

Docs Directory APIs SDKs

Store and retrieve context

AI applications need specialized storage for vector embeddings, conversation history, training data, and cached responses. Cloudflare Vectorize stores and queries embeddings for Retrieval Augmented Generation (RAG), D1 provides SQL storage for structured data, R2 stores documents and assets, and KV caches frequent responses at the edge.

Solutions

Vectorize

Vector database for storing and querying embeddings. Learn more about Vectorize.

Vector search - Store embeddings and find semantically similar content for Retrieval Augmented Generation (RAG) and recommendation features

D1

Serverless SQL database built on SQLite, with global read replication. Learn more about D1.

Structured storage - Structured Query Language (SQL) database for conversation history, user data, and application metadata

R2

S3-compatible object storage with zero egress fees. Learn more about R2.

Object storage - Store documents, training data, and generated assets with no egress fees

KV

Globally distributed key-value storage for low-latency reads. Learn more about KV.

Edge caching - Cache frequent AI responses at the edge to reduce inference costs and latency

Get started