TiDB
TiDB is built for agentic workloads that grow unpredictably, with ACID guarantees and native support for transactions, analytics, and vector search. No data silos. No noisy neighbors. No infrastructure ceiling. observed · 2026-08-28
Health v2 · maintenance only
99/100
- Activity 99
- Release rhythm 99
- Longevity 100
How is this computed?
round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10) — computed 2026-09-03. Adoption (stars, forks) is never an input.
- gap_med: 26.5
- age_days: 4014
- days_rel: 6
- days_push: 7
- n_releases_24m: 19
Adoption not part of the score
40473 stars · 6234 forks observed · 2026-08-28
What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded
TiDB is an open-source, cloud-native distributed SQL database written in Go, MySQL-compatible and designed for horizontal scalability, strong consistency, and high availability via Raft consensus. It supports HTAP workloads by combining row-based TiKV with columnar TiFlash storage, and now offers native vector search for AI agent workloads, with both self-managed and fully-managed cloud offerings.
Use cases
- scale a MySQL-compatible database horizontally without sharding
- run transactional and analytical queries on one database (HTAP)
- store and search vector embeddings for AI agent memory and context
- replace hundreds of sharded PostgreSQL or MySQL clusters
- build globally available, highly consistent backend storage for SaaS
- deploy a database on Kubernetes with automated operations
- migrate from MySQL + object storage to a unified database
When to choose
- you need MySQL compatibility with horizontal scale-out and no manual sharding
- you want ACID distributed transactions with strong consistency and automatic failover
- your workload mixes OLTP and OLAP and you want one unified database
- you are building agentic AI applications needing transactions plus vector search
- you need to scale to very large table counts and concurrent connections
When to avoid
- you need a small, single-node embedded database with minimal operational footprint
- your workload is purely analytical with no transactional requirements (a columnar warehouse may fit better)
- you depend on MySQL features TiDB does not support (triggers, some stored procedures, foreign keys in older versions)
- you want a lightweight key-value store rather than a full SQL database
- your team cannot operate a multi-component distributed cluster and you cannot use the managed cloud
Facets
service · maturity stable
database vector-database search-engine caching databases microservices large-language-models cloud-computing analytics cloud self-hosted go distributed-sql htap mysql-compatible raft horizontal-scaling acid-transactions tiflash tikv dbaas agentic-workloads vector-search mysql-protocol ai-agents linux macos kubernetes docker
9 sources
- readme: https://github.com/pingcap/tidb · fetched 2026-08-28 · 18dc3a29dd45
- homepage: https://www.tidb.io/ · fetched 2026-08-29 · bc5f1edb876f
- site_page: https://www.pingcap.com/about-us · fetched 2026-08-29 · a9c82ce741d7
- site_page: https://docs.pingcap.com/ · fetched 2026-08-29 · dcfbe4328796
- site_page: https://docs.pingcap.com · fetched 2026-08-29 · dcfbe4328796
- site_page: https://docs.pingcap.com/tidb/stable/faq-overview · fetched 2026-08-29 · 303b1e1e7087
- site_page: https://www.pingcap.com/pricing · fetched 2026-08-29 · f46024df64f3
- site_page: https://www.pingcap.com/integrations · fetched 2026-08-29 · 4b5ecd2cd833
- site_page: https://www.pingcap.com/press-releases-news · fetched 2026-08-29 · 7b06174a9992
Member repositories
| Repository | Role | Health v2 |
|---|---|---|
| pingcap/tidb | main | 99 |
| pingcap/docs-cn | docs | 76 |
| pingcap/parser | mirror | 23 |
| pingcap/tidb-operator | infra | 94 |
| pingcap/tiflash | backend | 100 |
For agents
Data as of 2026-08-30T08:39:29.467469+00:00 · Report a problem