Weaviate vs Qdrant
Weaviate bundles storage with hosted extras such as an embeddings service and a query agent, while Qdrant concentrates on the vector engine alone. Both publish their code under permissive licences and both are European companies, in Amsterdam and Berlin. The choice comes down to whether you want extras on one invoice or a smaller engine that is easy to move.
Sebastiaan SmitsFounderVendors can pay for more visibility on this website, for example a higher place in a ranking or more room on a page. Commercial relationships can also affect which products we feature and how prominently. Every vendor link is marked the same, whatever the arrangement behind it. How that works.
Choose Weaviate ifyou want embeddings, keyword filtering and reranking handled by one hosted service, and you accept extra usage meters next to the cluster fee.
Choose Qdrant ifyou want a single-purpose engine that runs on servers you control or can be managed inside your own infrastructure, and you will keep text and relational data elsewhere.
Weaviate and Qdrant side by side
| Weaviate | Qdrant | |
|---|---|---|
| Our vector databases rank | #2 of 12 | #1 of 12 |
| Established in | Netherlands | Germany |
| List price, 4 November 2026 | Free sandbox (100,000 objects); Flex billed pay-as-you-go; Premium as a prepaid contract through sales; embeddings priced per million tokens | Free-forever single-node cluster; Standard billed by usage with a calculator; Premium needs a minimum spend; self-hosting costs nothing in licence fees |
| Free tier | Yes | Yes |
| Open source | Yes | Yes |
| Self-hostable | — | Yes |
| Built for | Teams that want embeddings, hybrid search and storage handled in one hosted service | Teams that want a dedicated vector engine they can move between cloud and own servers |
| Not for | Buyers who need dedicated-cluster pricing before talking to sales | Teams that want one database for text, joins and vectors |
Both are reviewed in full in our vector databases ranking: Weaviate and Qdrant.
Where they differ
Weaviate 2 · Qdrant 2 · open 0- 01
What the free cluster holds
Weaviate's free sandbox is capped at 100,000 objects and 3 tenants, with a single collection, 1 GB memory and 10 GB disk. Qdrant's free tier is one node with 0.5 vCPU, 1 GB of RAM and 4 GB of disk, described as free forever. Weaviate states an object ceiling, and Qdrant states only hardware, so Weaviate is easier to size against.
Edge: Weaviate
- 02
Licence and moving away
Weaviate's repository lists BSD-3-Clause and runs locally in Docker. Qdrant's repository lists Apache-2.0, and its cloud runs the same engine, so leaving means pointing clients at your own server. Both are permissive, and neither locks the database itself. Qdrant adds Hybrid Cloud, which keeps managed clusters on your own infrastructure.
Edge: Qdrant
- 03
Extras around the database
Weaviate Embeddings is priced per million tokens for named models, and its Query Agent includes 1,000 free requests a month. Its README also names RAG, reranking, multi-tenancy, replication and RBAC. Qdrant describes a vector database and vector search engine and nothing broader. If you want fewer moving parts to assemble, Weaviate offers more built in.
Edge: Weaviate
- 04
Seeing the production price
Weaviate's Flex plan is pay-as-you-go on a shared cluster with 99.5% uptime, and Premium is a prepaid contract through sales. Qdrant's Standard tier is usage-based with dedicated resources and the same 99.5% figure, with a calculator on the pricing page. Neither gives a flat list price for scale, but Qdrant lets you compute one yourself.
Edge: Qdrant
When neither fits
If you want vectors inside a relational database you already run, or a larger distributed engine, look elsewhere. pgvector keeps vectors in Postgres, and Milvus is the other open-source engine in our vector database ranking.
Weaviate or Qdrant: questions
3 answersWhich one is easier to self-host?
Both publish their code, Weaviate under BSD-3-Clause with a Docker quickstart and Qdrant under Apache-2.0. Qdrant also offers Hybrid Cloud, where managed clusters run on your own infrastructure.
Where are the two companies based?
Weaviate B.V. is established in Amsterdam, the Netherlands. Qdrant Solutions GmbH is registered in Berlin, Germany.
Do I need another system for keyword search?
Weaviate's README describes combining vector search with keyword filtering. Qdrant describes itself as a vector search engine, so keyword-heavy search may need a second system.
Sources
Checked 4 November 2026- Weaviate pricing page: sandbox caps, Flex and contract tiers, embeddings per million tokens, Query Agent allowance.
- Weaviate source repository: licence listed as BSD-3-Clause.
- Weaviate README: product description with filtering, RAG, reranking and Docker quickstart.
- Weaviate privacy policy: Weaviate B.V. established in Amsterdam.
- Qdrant pricing page: free tier hardware, usage-based Standard, calculator, Hybrid Cloud, minimum spend tier.
- Qdrant on GitHub: Apache-2.0 licence and vector search engine description.
- Qdrant imprint: Qdrant Solutions GmbH in Berlin.