# zilliztech/VectorDBBench

Benchmark for vector databases.

Repository: https://github.com/zilliztech/VectorDBBench
Canonical: https://ross.abutalabs.com/products/vectordbbench
Homepage: https://zilliz.com/vector-database-benchmark-tool
Language: Python
License: MIT
License Family: permissive
Topics: benchmark, performance, vector-database, vector-search, vectordb, cost-effectiveness
Last push: 2026-08-26T08:13:20+00:00

## Health v2 (maintenance only)
Score: 91/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 84, longevity 86
- inputs: {"age_days": 1212, "days_push": 7, "days_rel": 110, "gap_med": 9, "n_releases_24m": 40}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1166, forks 431 (observed 2026-08-28T04:03:50.355390+00:00)

## What it is
VectorDBBench (VDBBench) is an open-source Python benchmark tool for comparing the performance and cost-effectiveness of vector databases and cloud vector services. It provides a visual interface, public datasets, and reproducible test scenarios including insertion, search, filtered search, and full-text (BM25) retrieval.

## Use cases
- benchmark vector database performance
- compare vector database cost-effectiveness
- reproduce vector database benchmark results
- test qps and recall of vector search systems
- evaluate cloud vector database services
- benchmark full text search bm25 retrieval
- choose the best vector database for my workload

## When to choose
- you need to compare vector databases like Milvus, Qdrant, Pinecone, or Elastic on QPS, latency, and recall
- you want reproducible benchmarks with real-world datasets such as SIFT, GIST, or Cohere
- you need cost-effectiveness reports for cloud vector services
- you want a GUI plus CLI to run and visualize benchmark results

## When to avoid
- you need to benchmark non-vector databases like SQL or document stores
- you only need a published leaderboard without running your own tests
- your system under test is not among the supported database clients

## Facets
- artifact type: cli-tool
- maturity: active
- function: benchmarking, vector-database, search-engine, data-visualization
- domain: databases, performance, developer-tools
- platform: python, cli, cross-platform
- tags: vector-search, vectordb, qps, recall, cost-effectiveness, leaderboard, load-testing, search

## Member repositories
- zilliztech/VectorDBBench (main) score 91

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:50.355390+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T06:29:19.535028+00:00, confidence not recorded.
  - readme: https://github.com/zilliztech/VectorDBBench (fetched 2026-08-28T04:03:50.355390+00:00, sha bb2afc4bdeab)
  - homepage: https://zilliz.com/vector-database-benchmark-tool (fetched 2026-08-29T12:35:21.564194+00:00, sha 40493a59070e)
- Data as of 2026-08-30T08:39:29.467469+00:00.