Compare Pinecone, Supabase pgvector, Qdrant, and Weaviate costs. Enter your vector count, dimensions, and query volume — see exact monthly prices.
Select provider · Enter vector count and queries · Results update live
Comparison for 1M vectors · 1536 dimensions · 100K queries/month.
| Provider | Free tier | Storage | Queries | 1M vec / 100K q |
|---|---|---|---|---|
| Supabase pgvectorBEST VALUE | 500MB + 2 projects | $0.125/GB | Included | ~$27 |
| Qdrant Cloud | 1GB storage | $9.00/GB | Included | ~$54 |
| Weaviate Cloud | 1 sandbox cluster | $0.50/GB | $10/M | ~$28 |
| Pinecone Serverless | 100K vectors | $0.33/GB | $16/M RU | ~$20 |
Qdrant on a $6/mo Hetzner VPS handles 1–5M vectors easily. Total cost: ~$6–10/mo vs $50+ on managed cloud. Requires ops knowledge.
Start building your vector search:
For under 5M vectors: Supabase pgvector at $25/month (queries included). For 50M+ vectors: self-hosted Qdrant on spot VMs at $150–300/month. Pinecone Serverless is cheapest for very small (under 100K) with the free tier.
Approximately $15–30/month depending on query volume. Storage ~$3 for 1536-dim vectors, plus $16 per million read units. At 100K queries/month expect ~$20.
For most production workloads yes. pgvector in Supabase or RDS supports HNSW indexing, IVFFlat, and handles millions of vectors well. Pinecone has better managed scaling and dedicated infrastructure for very high query rates.
Qdrant Cloud handles hundreds of millions of vectors. Self-hosted Qdrant on a 16GB RAM server can store ~10M 1536-dim vectors in memory, or 50M+ with memory-mapped storage (slower queries).
Dimensions are the size of each embedding vector. OpenAI text-embedding-3-small produces 1536-dim vectors. Larger dimensions = more storage and cost. You can reduce OpenAI embeddings to 256 or 512 dimensions with minimal quality loss.