Product / UX / Agent tools & protocols

Qdrant

High-performance vector database and search engine built in Rust, designed for the next generation of AI applications. Stores vectors with rich payloads and supports advanced filtering, making it ideal for RAG retrieval pipelines that agents call to access knowledge bases. 31K+ GitHub stars, available as open source self-hosted or managed cloud. The leading Rust-native vector DB, competing with Milvus and Pinecone.

High-performance Rust vector database / vector search engine (dense, sparse, multi-vector, hybrid search, quantization). Apache-2.0 core + Qdrant Cloud managed tier (free tier available). v1.18.2 released 4 Jun 2026; verified live June 2026. Calibrated against frozen anchor Milvus (O5 open_source/A3/C4).

Openness

4 high confidence

4.0

license: Apache-2.0(OSI, engine core)
source: public(Rust)
managed-tier: Qdrant Cloud (hosted, paid + free tier)
core-gated: gated

Engine core is Apache-2.0 OSI, but a managed Qdrant Cloud tier sits on top; recipe explicitly classes Qdrant Cloud as open_core, so scored 4 rather than the pure-OSS 5 the Milvus anchor carries.

https://github.com/qdrant/qdrant/blob/master/LICENSE recorded 2026-06-04
Apache-2.0 license text
https://github.com/qdrant/qdrant recorded 2026-06-04
v1.18.2 (4 Jun 2026); Qdrant Cloud managed tier with free option

Adoption

5 high confidence

5.0

Re-scored 4->5. qdrant-client 23.06M PyPI downloads in the last 30 days (verified on pypistats; the prior note already reported ~26.1M but mislabeled the reach as 1M-10M). The map's prevailing usage_volume level-5 floor is >10M/mo (cf. pydantic-ai ~31M, langgraph ~58M at level 5); Qdrant's client volume clears it. It remains one above the Milvus anchor (A3) on download volume, a leading open vector DB for RAG/agent retrieval.

https://pypistats.org/api/packages/qdrant-client/recent recorded 2026-06-25
last_month downloads = 23,061,488

Capability

4 medium confidence

4.0

Top-tier ANN engine on RPS/latency; tied with the Milvus anchor at C4 (frontier vector-DB capability). Headline comparison is vendor-run (qdrant.tech/benchmarks); ANN-Benchmarks corroborates inclusion but its interactive plots did not yield a clean ranking number this run, hence medium confidence.

https://ann-benchmarks.com/ recorded 2026-06-04
Qdrant included among 37 benchmarked ANN algorithms
https://qdrant.tech/benchmarks/ recorded 2026-06-04
Qdrant highest RPS/lowest latency vs Milvus/Weaviate/Elasticsearch/Redis (vendor benchmark)

Unchanged since 2026-07-30 (last edited, not re-checked)