Pyyan / Compare / Redis vs pgvector vs Qdrant vs Pinecone vs Milvus
Redis vs pgvector vs Qdrant vs Pinecone vs Milvus
5 of 5Vector Databases · verified 13 Aug 2026
| Specification | Redis | pgvector | Qdrant | Pinecone | Milvus |
|---|---|---|---|---|---|
| Summary | In-memory, so very fast and RAM-bound. | Vectors inside the database you already run. | The fastest of the purpose-built stores. | Managed, and the smoothest to operate. | Built for billions of vectors. |
| Index | HNSW, FLAT | HNSW, IVFFlat | HNSW | Proprietary | HNSW, IVF, DiskANN, GPU |
| Hybrid search | Yes | Yes, with SQL | Yes | Yes | Yes |
| Hosting | Self-host and managed | Self-host or any managed Postgres | Self-host and managed | Managed only | Self-host and managed |
| Licence | RSALv2 / SSPL | PostgreSQL | Apache 2.0 | Proprietary | Apache 2.0 |
| p50 latency | 5ms | ~15ms | 4ms | <10ms | ~10ms |
| Category | Vector Databases | Vector Databases | Vector Databases | Vector Databases | Vector Databases |
| Official | Redis ↗ | PostgreSQL ↗ | Qdrant ↗ | Pinecone ↗ | Zilliz ↗ |
Highlighted rows are where these differ.
Redis
- Practical to roughly 10 to 100 million vectors
- You are paying for RAM, which sets the ceiling
Best for caches and hot vector sets.
pgvector
- One database for vectors, rows and joins
- The honest default: reach for something else only when this stops working
Best for almost everyone, until scale says otherwise.
Qdrant
- Written in Rust; 10 to 25% faster than Weaviate or Milvus on common workloads
- Strong filtering alongside vector search
- Self-host or managed, same engine
Best for latency-sensitive retrieval.
Pinecone
- Sub-10ms p50 with nothing to maintain
- Costs draw scrutiny at scale; Notion moved away and cut spend ~60%
Best for teams who do not want to run a database.
Milvus
- Distributed architecture with GPU index support
- Heavier to operate than the alternatives
Best for very large corpora.