Qdrant vs Weaviate vs Milvus vs Chroma
A side-by-side look at scores, pricing and features — with RECATOOLS' ASEAN-aware verdict for each.
Qdrant
Rust vector search engine built for filtering-heavy AI workloads
Visit
|
Weaviate
Open-source vector database with hybrid search built in
Visit
|
Milvus
Open-source vector database for billion-scale search, now lake-native
Visit
|
Chroma
Simple open-source vector and full-text search, from laptop to cloud
Visit
|
|
|---|---|---|---|---|
| RECATOOLS Score | 8.7 / 10 | 8 / 10 | 8.3 / 10 | 7.5 / 10 |
| Capability | ||||
| Value for money | ||||
| Ease of use | ||||
| ASEAN readiness | ||||
| API quality | ||||
| Pricing | Open Source | Open Source | Open Source | Open Source |
| Free tier | Free forever cluster: 1 node, 0.5 vCPU, 1GB RAM, 4GB disk (~1M 768-dim vectors per docs), no card; suspended after 1 week idle. Self-hosted engine free (Apache-2.0) | Always-free Weaviate Cloud sandbox (100,000 objects, 1 GB memory, 10 GB disk, 1 collection); BSD-3-Clause core free to self-host | Milvus is free to self-host under Apache-2.0. Zilliz Cloud Free cluster: 5GB storage, 2.5M vCUs/month, up to 5 collections, community support | Open-source core free under Apache 2.0. Chroma Cloud Starter $0 + usage with $5 free credits, 10 databases, 10 team members |
| Paid from | Standard: usage-based, billed hourly on vCPU, RAM, disk and backups | Flex from $45/mo pay-as-you-go; Premium prepaid contract, minimum from $400/mo | Milvus free (Apache-2.0); Zilliz Cloud Dedicated from $126/mo (Standard) | Chroma Cloud Team $250/mo + usage |
| Has API | ✓ | ✓ | ✓ | ✓ |
| Open source | ✓ | ✓ | ✓ | ✓ |
| Free to use | ✓ | ✓ | ✓ | ✓ |
| Users | — | — | — | — |
| Founded | 2021 | 2019 | 2019 | 2022 |
| Maker | Independent (Qdrant Solutions GmbH) | — | Zilliz (LF AI & Data Foundation project) | Independent (Chroma Inc.) |
| Verdict | Qdrant is one of the strongest open-source vector databases, especially where search has to be combined with complex filters, and its free cloud cluster and permissive licence make it easy to start. Its performance claim... |
Weaviate pairs vector similarity with BM25 keyword search in a single query, and its built-in vectorizer modules mean you are not always wiring your own embedding pipeline. G2 reviewers rate it 4.4 out of 5 across 49 rev... |
Milvus is one of the most capable open-source vector databases, scaling from a laptop library to billions of vectors, and version 3.0's lake-native search is a genuine step forward. Running it distributed takes real oper... |
Chroma remains the friendliest way to add vector search to an AI app, with a smooth path from an embedded library to a managed cloud that now has enterprise features such as private networking and an EU region. Its own f... |
| Full review → | Directory listing → | Directory listing → | Directory listing → |
Comparisons cover up to 4 tools. Scores are RECATOOLS editorial assessments; verify current pricing on each vendor's site.