#1919th of 22 in Vector databases
attu
Management UI for Milvus clusters with data browsing, search and an AI agent
- Stars
- 3.2k
- License
- custom license
- Last commit
- Sep 2026
- Last release
- Sep 2026
- Language
- Shell
Missed the latest GitHub check; these facts may be older.
Zilliz is hiring: 12 open roles
Overview
Attu is a management console for Milvus vector databases. One instance connects to several Milvus clusters or Zilliz Cloud, and lets you browse collections, edit data, run vector and BM25 searches, manage backups and RBAC, and view Prometheus metrics. A built-in agent with 50+ tools performs Milvus operations from chat, using OpenAI, Claude, DeepSeek, Gemini, OpenRouter or a custom endpoint.
Who it is for: Teams running Milvus who want a GUI for admin, search and backups
Strengths
- Multiple Milvus clusters managed from one instance, each with its own workspace
- Ships as Docker image, Linux server package, or desktop app for macOS, Linux, Windows
- Local login, audit logs of write operations, TLS/mTLS and SSRF protection built in
- Backup and restore to S3, MinIO, GCS or Azure Blob, plus CSV, JSON, Parquet import/export
Weaknesses
- Supports only Milvus 2.6.0+ and 3.x plus Zilliz Cloud; 2.5.x and earlier are unsupported
- Works only with Milvus; no other vector stores
- Desktop build is single-user and has no multi-user login
- Desktop SSRF rules block local and LAN services unless allowlisted via environment variable
What it needs
- no GPU
- Needs Milvus 2.6.0+ or 3.x, or Zilliz Cloud, SQLite (bundled), Node.js 24.x (standalone server package)
- Models: OpenAI, Anthropic Claude, DeepSeek, Google Gemini, OpenRouter
- port 3000
Also in Vector databases
See all 22| Rank | Project | Score |
|---|---|---|
| 1 | MilvusDistributed vector database with dense, sparse and hybrid search at scale | 90 out of 100 |
| 2 | QdrantRust vector database with payload filtering, REST and gRPC | 85 out of 100 |
| 3 | MeilisearchRust search engine with typo tolerance, filtering and hybrid semantic search | 84 out of 100 |
| #4 | faissC++ library for similarity search and clustering of dense vectors | 81 out of 100 |
| #5 | leannVector index that recomputes embeddings on demand to cut storage | 79 out of 100 |