Vald
Distributed vector search is provided by Vald, an engine designed to scale across deployments.
About Vald
Vald is an open-source, highly scalable distributed vector search engine built on cloud-native architecture, designed for fast approximate nearest neighbor (ANN) dense vector search. It supports automatic vector indexing, index backup, and horizontal scaling to handle search across billions of feature vectors, and uses the NGT ANN algorithm for neighbor searches. The engine can process vectorized data from images, audio, video, text, binary and other object types for use cases including recognition, recommendation, detection, grammar checking and real-time translation. Deployment requires Kubernetes 1.19 or later and AVX2 instruction support for the Vald Agent NGT component.
Key features
- Automatic vector indexing and index backup
- Horizontal scaling for billions of feature vectors
- Processes vectorized image, audio, video, text, binary data
- Uses NGT ANN algorithm for neighbor search
- Cloud-native architecture for Kubernetes deployments
- Customizable feature set for varied use cases
Best for: Teams building large-scale vector search systems for recognition, recommendation, detection, translation or similar use cases.
Good for
- Supports horizontal scaling to search billions of feature vectors
- Uses the NGT ANN algorithm for neighbor searches
- Open source under the Apache 2.0 license
Watch out for
- Requires Kubernetes 1.19 or later for deployment
- Requires AVX2 instruction support for Vald Agent NGT
- Requires input data to be pre-converted to vector format
Vald at a glance
- Website
- github.com/vdaas/vald
- Category
- Development › APIs & Databases
- Type
- Open-source project
- Pricing
- Not confirmed. Check the site
- Open source
- Yes · Source code
- Language
- English
- Last checked
- Oct 5, 2026 · Online when checked
- Listed
- Oct 5, 2026
Guides for APIs & Databases
Vald alternatives
Tools and sites you could use instead, ranked by how closely they match.
Vector similarity search is provided by Qdrant, an open-source search engine written in Rust.
AI search and serving structured data are provided by Vespa, an open-source software platform.
Open source
24 alternatives
ParadeDB adds full text search, vector retrieval, and aggregations to PostgreSQL.
Open source
24 alternatives
PostgreSQL vector search can use pgvecto.rs, an extension described as scalable and low latency.
Open source
24 alternatives
Vector search and storage are the focus of VectorDB, a Python vector database project.
Open source
24 alternatives
Distributed search and analytics are provided by Elasticsearch, an open-source engine for indexed data.
Free trial
13 alternativesPopular in APIs & Databases
Well-known listings from the same topic.
API3api3.orgTwiliotwilio.comGelgeldata.comDB Browsersqlitebrowser.orgPocketBasepocketbase.ioBrunousebruno.com
Vald FAQ
What type of search engine is Vald?
Vald is a highly scalable distributed fast approximate nearest neighbor (ANN) dense vector search engine.
What are the system requirements for deploying Vald?
Deployment requires Kubernetes 1.19 or later and AVX2 instruction support for the Vald Agent NGT component.
What use cases is Vald suitable for?
Vald can be used for recognition, recommendation, detection, grammar checking, real-time translation and other vector-based tasks.
Is something wrong or out of date?
Tell us and an editor will check it.
Own this website?
Claim this listing
Claim the listing and suggest updates yourself. An editor checks every claim first.