Milvus vs lancedb
Two of the top projects in Vector databases, side by side: who leads on each score part and who each one suits, then setup, license, activity and what each review found.
Milvus
Distributed vector database with dense, sparse and hybrid search at scale
lancedb
Embedded vector database built on the Lance columnar format
Verdict
Milvus ranks higher, 83 to 69. Of the 5 score parts we measure for both, Milvus leads on 2, lancedb on 1 and 2 are even.
- AdoptionMilvus leads by 56 points, 90 to 34: a wide gap.
- FreshnessEven at 100.
- Maintenancelancedb leads by 4 points, 88 to 84: close to even.
- Easy to runEven at 67.
- Agent-readyMilvus leads by 40 points, 70 to 30: a wide gap.
Pick Milvus if you are
Teams needing billion-scale vector search on Kubernetes
- Index types HNSW, IVF, FLAT, SCANN, DiskANN plus GPU CAGRA
- Dense, sparse (BM25, SPLADE, BGE-M3) and hybrid search in one collection
Pick lancedb if you are
Developers needing an embedded vector store for multimodal AI apps
- Runs embedded and locally, with no separate server required per the README
- Combines vector search, full-text search and SQL in one store
| What we compare | Milvus | lancedb |
|---|---|---|
| Score parts, out of 100 | ||
| Adoption | 90, widely used | 34, known |
| Freshness | 100, active | 100, active |
| Maintenance | 84, healthy | 88, healthy |
| Easy to run | 67, easy | 67, easy |
| Agent-ready | 70, partly | 30, minimal |
| Facts from GitHub and the README | ||
| Stars | 46.5k | 11.7k |
| License | Apache-2.0 (permissive) | Apache-2.0 (permissive) |
| Last commit | Oct 2026 | Oct 2026 |
| Last release | Oct 2026 | Oct 2026 |
| Language | Go | Rust |
| Docker | Yes | Yes |
| GPU | Optional | Optional |
| arm64 or Apple Silicon | Mentioned | Mentioned |
Milvus
Milvus is a distributed vector database (Go and C++, LF AI & Data Foundation) that separates compute and storage on Kubernetes, with a Standalone Docker mode and pip-installable Milvus Lite. It offers HNSW, IVF, FLAT, SCANN and DiskANN indexes, GPU CAGRA, sparse BM25 and learned-sparse vectors for hybrid search, metadata filtering, multi-tenancy, hot/cold storage, auth, TLS and RBAC.
Who it is for: teams needing billion-scale vector search on Kubernetes
Strengths
- Index types HNSW, IVF, FLAT, SCANN, DiskANN plus GPU CAGRA
- Dense, sparse (BM25, SPLADE, BGE-M3) and hybrid search in one collection
- Multi-tenancy at database, collection, partition or partition-key level
- Mandatory auth, TLS and RBAC; Milvus Lite via pip for local dev
Weaknesses
- Distributed mode is Kubernetes-native with several microservices to operate
- No port, RAM or Docker command in the README; install lives in docs
- Zilliz is the major contributor and promotes its managed cloud
- Source build needs Go 1.21+, CMake, GCC 11+ and Python 3.8 to 3.11
- GPU optional
- Compose
- Compose runs Milvus, MinIO
- Models: any embedding model or service; pymilvus[model] wraps embedding and reranking models
lancedb
LanceDB is a vector database written in Rust on top of the Lance columnar format. It stores vectors, metadata and multimodal data (text, images, video, point clouds) and supports vector similarity search, full-text search and SQL filtering. It runs locally or in your own cloud, with Python, TypeScript and Rust SDKs plus a REST API.
Who it is for: Developers needing an embedded vector store for multimodal AI apps
Strengths
- Runs embedded and locally, with no separate server required per the README
- Combines vector search, full-text search and SQL in one store
- Automatic data versioning and zero-copy access without extra infrastructure
- Python, TypeScript and Rust SDKs; integrates with LangChain, LlamaIndex, Pandas, Polars, DuckDB
Weaknesses
- README gives no benchmarks or hardware figures to back the billion-vector claims
- Managed Cloud and Enterprise tiers exist; feature split with open source is not detailed
- README has little setup detail and defers to external docs
- RAM and GPU requirements are not stated
- GPU optional
- Docker + Compose
More in Vector databases
Building with these?
We help teams pick an open-source AI stack and run it in production. Client work never changes a rank.