PostgreSQL Vector Search
Configure half-precision vectors, HNSW indexes, filtered retrieval and vector-search tuning.
- Purpose
- Vector similarity search
- Author / publisher
- Tiger Data
- License
- Apache-2.0 · License guidance
- Version
- b236d3583fb5
- Access
- Free download. External tools and services may have separate costs.
- Declared clients
- Claude Code, Codex, Cursor
Source & package
5 files · 0 script/code files. License file detected.
Current package SHA-256
827827a5063c36b5b1e920e853c0005812e93f39ba20daf6252703e19038414cTools & requirements
PostgreSQL with pgvector 0.8.0+ for all features, compatible binaries and installation rights. Generate embeddings separately with a suitable model; optional vectorscale is separate.
Apache-2.0 material; separate hosting, tooling, model access and services may have costs. Full skill/agent runtime evaluation has not been performed. Capacity, recall, speed and memory tables are author estimates, not measured guarantees. Use nonzero vectors of matching dimensions from the same embedding model. Review large maintenance_work_mem values against your limits. EXPLAIN ANALYZE executes queries. Optional maintenance/DDL is not read-only. No model, credentials or binaries bundled.
Evidence & limits
Source review checked 2026-10-03. Full skill/agent runtime evaluation has not been performed.
1 current outcome record. Read the actual environment, result and limitations.