seekdb

The AI-Native Search Database. Best for agent storage, it unifies vector, text, structured, and semi-structured data into a single engine. This all-in-one datab

What is seekdb?

The AI-Native Search Database. Best for agent storage, it unifies vector, text, structured, and semi-structured data into a single engine. This all-in-one database makes agents smarter, easier to run, and more stable.

Overview of seekdb

What it is

seekdb is an AI-native search database that unifies vector, text, structured, and semi-structured data into a single engine. It is MySQL-compatible and can run as an embedded library, a single-node server, or in a distributed cluster. The database is designed for agent storage, making agents smarter, easier to run, and more stable.

Key features

  • Streaming write + concurrent search: Achieves 1,523 QPS with 21.7 ms concurrent P99, claimed to be 10.7× the QPS of Milvus and 3× Elasticsearch, with flat P99 under concurrency.
  • Copy-on-Write sandboxes: FORK DATABASE snapshots an entire database in seconds without data copy; MERGE TABLE commits work back, DROP DATABASE discards it.
  • Hybrid search in single SQL: Combines vector, full-text, and scalar filters in one execution plan, avoiding N+1 client-side merging.
  • MySQL-compatible, ACID, embeddable: Built on OceanBase SQL engine, supports full ACID, real-time writes, and works with LangChain, LlamaIndex, Dify.
  • Async index pipeline (Change Stream): Decouples DML from index build; two-level HNSW (incremental + snapshot) makes newly-written vectors immediately searchable.

Use cases

  • Agentic AI: Memory, sandbox, and state store for personal assistants, enterprise automation, vertical agents, and agent platforms.
  • RAG & Knowledge Retrieval: Vector + full-text + scalar filters with multi-level access control for enterprise QA, customer support, knowledge bases.
  • Semantic Search: Embedding-based search across text, images, and other modalities for product search, text-to-image, image-to-product.
  • AI-Assisted Coding: Semantic code search, multi-project isolation, time-travel queries for IDE plugins and code agents.
  • Enterprise Application Intelligence: MySQL-compatible AI layer for legacy systems with row/column hybrid storage.
  • On-Device & Edge AI: Embedded/micro-server modes for resource-constrained devices like in-vehicle systems, AI education, companion robots.

Getting started

Installation options from the README:bash# Python (recommended for AI/ML)pip install -U pyseekdb# Docker (quick testing)docker run -d --name seekdb -p 2881:2881 -p 2886:2886 -v ./data:/var/lib/oceanbase oceanbase/seekdb:latest# Cloud (zero install, free for 7 days)curl -X POST https://d0.seekdb.ai/api/v1/instances# Binary (Linux one-line install)curl -fsSL https://obportal.s3.ap-southeast-1.amazonaws.com/download-center/opensource/seekdb/seekdb_install.sh | bash# Binary (macOS Homebrew)brew tap oceanbase/seekdb && brew install seekdb

License and pricing

Licensed under Apache-2.0. The project is open source with no stated pricing.

Based on the project's public README and repository metadata.

Add this badge to your README

Featured on YunoDB Badge styles & help

Maintain this project? Show that it is featured on YunoDB.

Markdown[![Featured on YunoDB](https://yunodb.com/badge/oceanbaseseekdb.svg)](https://yunodb.com/tools/oceanbaseseekdb?utm_source=github&utm_medium=badge)
HTML<a href="https://yunodb.com/tools/oceanbaseseekdb?utm_source=github&utm_medium=badge"><img src="https://yunodb.com/badge/oceanbaseseekdb.svg" alt="Featured on YunoDB" height="20"></a>