oceanbase/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.

What it solves

seekdb is a state store designed specifically for AI agents that requires continuous memory writes and immediate retrieval. It solves the performance bottlenecks found in traditional vector databases when handling high-concurrency streaming writes and concurrent searches, and provides a mechanism for agents to experiment with state without risking the main database.

How it works

  • Async Index Pipeline: It uses a Change Stream to decouple data modification (DML) from index building, allowing writes to commit immediately while an asynchronous process updates a two-level HNSW index (incremental and snapshot).
  • Copy-on-Write (COW) Sandboxes: The FORK DATABASE command creates near-instant snapshots of the entire database without copying data, allowing agents to test changes in a sandbox that can later be merged back or discarded.
  • Hybrid Search: It integrates vector similarity, full-text search, and scalar filtering into a single SQL execution plan, eliminating the need for client-side merging.
  • MySQL Compatibility: Built on the OceanBase SQL engine, it supports the MySQL protocol and ACID transactions, making it compatible with existing tools like LangChain and LlamaIndex.

Who it’s for

  • AI Agent Developers: Those building personal assistants, enterprise automation, or agent platforms that need robust episodic memory and state management.
  • RAG Developers: Engineers building knowledge retrieval systems that require hybrid search (vector + text).
  • Edge AI Developers: Those needing an embeddable database for resource-constrained devices or companion robots.

Highlights

  • High Throughput: Achieves 1,523 QPS for streaming write+search workloads.
  • Instant Sandboxing: Kernel-level COW for rapid database forking and merging.
  • Unified Querying: Vector, full-text, and scalar filters in one SQL query.
  • Flexible Deployment: Available as an embedded library, a single-node server, or a distributed cluster.

Related

  • Project
    alibaba/AliSQL

    An open-source MySQL branch that integrates DuckDB for high-performance analytics and native HNSW vector search for AI applications.

  • Project
    lancedb/lancedb

    A multimodal AI lakehouse and vector database built on the Lance columnar format for fast, scalable search over petabytes of vectors and multimodal data.

  • Project
    justrach/codedb

    codedb is a zero‑dependency, Zig‑based server that builds rich indexes (trigram, word, symbol, call‑graph) of a local codebase and exposes them via the Model Context Protocol (JSON‑RPC over stdio). AI coding assistants (Claude Code, Cursor, Gemini CLI, etc.) can query for file trees, outlines, symbol definitions, callers, and full‑text search in sub‑millisecond latency, dramatically reducing token usage. It runs as an MCP daemon (`codedb mcp`) or an HTTP server, updates incrementally via a file watcher, and ships signed binaries for macOS, Linux, and Windows.

  • Project
    MariaDB/server

    An open-source relational database that provides MySQL compatibility and advanced features like native vector search and pluggable storage engines.

  • Project
    pingcap/tidb

    An open-source, cloud-native distributed SQL database that provides horizontal scalability, high availability, and MySQL compatibility.