VectorDBBench
Open-source vector database benchmarking tool testing databases across production-critical scenarios including static collection, filtering, and streaming cases with modern embedding model datasets.
About this tool
Overview
VDBBench is an open-source vector database benchmarking tool designed for comparing and evaluating vector databases. It's designed for users who require high-performance data storage and retrieval systems.
Database Support
Supports testing of:
- Milvus
- Zilliz Cloud
- Elasticsearch
- Qdrant Cloud
- Weaviate Cloud
- PgVector
Key Features
Real-World Testing
VDBBench is the only benchmark tool that tests vector databases across the complete spectrum of production-critical scenarios:
- Static collection tests
- Filtering scenarios
- Streaming cases
Comparison Capabilities
Offers the ability to select and compare results from multiple tests simultaneously, making it easy to understand relative performance across different vector databases.
Modern Datasets
VDBBench uses vectors generated from state-of-the-art embedding models that power today's AI applications, moving beyond legacy datasets like SIFT and GloVe.
Benchmark Metrics
VDBBench measures:
- p99 serial search latency
- Maximum concurrent QPS at 90% data capacity
- Performance while insertion workload remains active
Availability
- GitHub: https://github.com/zilliztech/VectorDBBench
- PyPI: Can be installed via pip as vectordb-bench
- Leaderboard: Public results viewable at the VDBBench Leaderboard
Use Cases
- Evaluating vector databases for production deployments
- Comparing performance across different solutions
- Validating POC requirements match production capabilities
- Performance regression testing
Pricing
Free and open-source.
Loading more......
Information
Categories
Tags
Similar Products
6 result(s)