HomeProjectsArticlesPK BoardAbout
Explore Projects

ClickHouse

Active
GitHub C++ Apache-2.0

Description

ClickHouse is an open-source OLAP columnar database with native vector search and SQL analytics, usable as an analytical retrieval engine for agents and RAG.

Key Features

  • Columnar OLAP — billion-row queries with second-level latency
  • Native SQL — ANSI SQL plus rich analytical functions
  • Vector search — built-in ANN indexes, vector + SQL hybrid queries
  • High-throughput writes — millions of inserts per second per node
  • Rich ecosystem — JDBC/ODBC, HTTP and native protocol
  • Cloud-native deployment — sharding, replicas, cross-DC replication

Use Cases

💡 Storing user behavior logs and observability events for agent platforms
💡 Backing RAG with SQL access to massive document and vector stores
💡 Real-time analytics on business and AI usage metrics
💡 A high-performance replacement for TimescaleDB / ElasticSearch

Strengths & Limitations

✅ Strengths

  • • Actively maintained, recent updates
  • • High community interest (49.3k stars)
  • • Permissive open-source license (Apache-2.0)

⚠️ Limitations

  • • High issue backlog (6.9k open issues)

Tags

olap columnar-database vector-search sql analytics

Categories

📚 RAG Tools ⚡ Agent Tools

Quick Start

curl -O https://clickhouse.com/clickhouse
chmod +x clickhouse
./clickhouse server

# In another shell
./clickhouse client
Visit GitHub Visit Website

Related Projects

Apache Doris

15.8k · Java
Active A

Apache Doris is a real-time analytics MPP database for AI agent workloads, delivering millisecond queries with hybrid search.

olapreal-time-analyticsvector-search +3
  • · Real-time analytics — Columnar MPP architecture delivers sub-second aggregation queries at scale
  • · Vector search — Built-in ANN index supports hybrid structured and unstructured search
  • · MySQL compatible — Standard MySQL protocol and SQL syntax
View full details → View on GitHub

Meilisearch

59.0k · Rust
Active A

Meilisearch is a lightweight open-source search engine with typo tolerance, native vector search and instant indexing, commonly used as the retrieval layer for agents and RAG.

search-enginefull-text-searchvector-search +2
  • · Typo-tolerant — forgiving search out of the box
  • · Instant indexing — new or updated docs are searchable within milliseconds
  • · Native vector search — built-in ANN vector index, hybrid with full-text
View full details → View on GitHub