Qdrant
Qdrant is an open-source vector search engine plus a managed Cloud, not an unpriced modeling tool. Rechecked 2026-08-17: qdrant.tech/pricing sells Cloud Free, Standard usage-based, and Premium with a minimum spend, plus Hybrid and Private Cloud contact rows. github.com/qdrant/qdrant had 34,018 stars, license Apache-2.0, homepage qdrant.tech. Drop leftover website qdrant.io and any copy that never printed a Cloud tier.
Compare Pinecone if you wanted a fully managed vector service with public $20/$50/$500 rows, Milvus if you wanted another OSS engine, and Supabase if you wanted Postgres instead.
Key Features
- OSS plus Cloud: Self-host the Apache-2.0 engine, or run Free / Standard / Premium on Qdrant Cloud. Hybrid Cloud and Private Cloud stay on your infrastructure.
- Free Cloud is a small node: FAQ says Free includes 1 GB RAM and 4 GB disk (page also lists 0.5 vCPU / 1 GB RAM / 4 GB disk). Exceed that and upgrade to Standard.
- Standard is usage-based: Dedicated resources, HA, backups, 99.5% SLA. Premium needs a minimum spend and adds SSO, private VPC links, 99.9% SLA.
- Serverless is coming soon: The nav still says Qdrant Serverless (Coming soon). Do not sell it as shipped.
Use Cases
- People who will stay on OSS or Cloud Free for prototypes.
- People who will pay Standard for a dedicated cluster.
- Enterprises that will talk to sales for Premium / Hybrid / Private.
Limitation: We did not invent a Standard dollar-per-hour. Use the live pricing calculator.
Pricing
qdrant.tech/pricing on 2026-08-17.
| Plan | Price | Notes from that page |
|---|---|---|
| OSS | $0 | Apache-2.0. 34,018 stars. Self-host. |
| Cloud Free | Free forever | 0.5 vCPU / 1 GB RAM / 4 GB disk. Testing and prototypes. |
| Cloud Standard | Usage-based | Dedicated resources, HA, 99.5% SLA. Marketplace deploy. |
| Cloud Premium | Minimum spend | SSO, private VPC links, 99.9% SLA. Talk to engineers. |
| Hybrid / Private Cloud | Contact | Run on your infra or isolated/air-gapped setups. |
If a leftover blog still points at qdrant.io with no Cloud tiers, treat this table as the source.
Getting Started
- Open qdrant.tech/pricing and start on Free or OSS.
- Recheck the calculator before you assume Standard is a $20 seat.
- Recheck Pinecone if you wanted a public $20 Builder row.
- Do not sell Serverless as GA.
Frequently Asked Questions
Is the engine free?
Yes. OSS is Apache-2.0. Cloud also has a free forever node.
What is the official site?
qdrant.tech. GitHub homepage matches that.
Same as Pinecone?
No. Pinecone's public rows are Free / $20 / $50 min / $500 min. Qdrant Standard is usage-based.
Alternatives
- Pinecone: Managed vector DB with a $20 Builder row.
- Milvus: Another open vector database.
- Supabase: Postgres BaaS, not a dedicated vector engine.
Tips
- Quote OSS $0 / Cloud Free / Standard usage-based / Premium min spend.
- Point website at qdrant.tech.
- Recheck the calculator before you invent a Standard dollar.
Conclusion
Qdrant is an Apache-2.0 vector engine with a priced Cloud (Free, usage-based Standard, Premium min spend), not an unpriced modeling tool. Start at qdrant.tech/pricing, then decide whether Pinecone already covers the managed seat you need.
Comments
No comments yet. Be the first to comment!
Related Tools
Pinecone
www.pinecone.io
Managed vector DB. Rechecked: Starter Free; Builder $20; Standard $50 min; Enterprise $500 min. Not an unpriced ML toolkit.
Milvus
milvus.io
Open distributed vector DB. Rechecked: Apache-2.0, 45,663 stars. Cloud is Zilliz; not a trillion-vector census.
Vespa
vespa.ai
OSS search+vector engine. Rechecked: Apache-2.0, 7,052 stars. Cloud has a free trial; unit table exists. Not an unpriced toolkit.
Related Insights

Grok Bot and Hermes Bot: one person finally gets a think tank and a secretariat
Grok Bot now ships with Cursor Pro+. Hermes Bot runs on a VPS. They are not smarter chat boxes. The think tank advises, the secretariat executes, and you still make the call.
Hook OpenCode Go into Codex on Windows. Do Not Open a Second Toolkit.
A ChatGPT-signed Codex desktop app still shows mostly GPT in the picker. On Windows, enable only OpenCode Go and the Grok, GLM, Kimi, DeepSeek, and MiniMax models you already pay for appear in the same selector. Keys stay local. Native GPT stays put.
What Locks Codex Is the Picker, Not the Models
You already pay for OpenCode Go, Grok, and Z.ai, but the Codex picker still shows mostly GPT. The community project codex-router does not teach another install ritual. It puts subscriptions you already bought back into the selector, keeps keys on the machine, and leaves native GPT alone.