kitio.aiAdded Nov 5, 2026
Vald
FreeOpen-source distributed vector search engine for billions of vectors
What is Vald?
Vald is a cloud-native approximate nearest neighbor search engine for dense vectors, built to scale horizontally across many machines. It suits engineering teams running similarity search over very large embedding collections on Kubernetes.
Vald uses the NGT algorithm and spreads its index across many agents, so indexing runs asynchronously without stopping searches. Each index is replicated across agents and rebalanced automatically if one goes down, and backups can go to object storage or persistent volumes.
Ingress and egress filters can be customized around its gRPC interface, and vector dimensions and replica counts are configurable. Official SDKs cover Go, Java, Node.js and Python.
Features
- Distributed ANN search
- Asynchronous auto-indexing
- Index replication
- Automatic index backup
- Custom gRPC filters
- Multi-language SDKs
Tags
Releases
+ Submit new release- Listed
Vald was added to the directory.
Organization
vdaas
Listed by kitio.ai · unverified owner
Pricing
- Pricing model
- Free
- Paid options from
- 100% free
Sponsored
Your AI tool here
A Featured launch puts your tool in this spot and on the home page for a week.
Top Vald alternatives
View allModels & APIs
Search engine and vector database for AI apps and agents
Models & APIs
Open-source vector database with hybrid search and built-in vectorizers
Models & APIs
Open-source vector database built for billion-scale search
If you liked Vald
Open-source embedding database for building RAG apps quickly
Freemium · $250/moModels & APIsMade in United States- MLServerNew
Open-source Python server for serving ML models over REST and gRPC
FreeModels & APIsMade in United Kingdom - PolyaxonNew
Open-source control plane for ML training, serving and agents on Kubernetes
Freemium · $555/moModels & APIs Kubernetes-native MCP and AI gateway platform for enterprise agents
FreemiumModels & APIsMade in United States
Pros and cons
Pros
- Free and open source
- Scales to billions of vectors
- Keeps serving while indexing
Cons
- Needs Kubernetes to run
- No managed cloud offering
- More setup than hosted vector databases
Prompts & results
No prompts yet. Add your own prompt and output to help others use this tool.
Share a prompt
Reviews
Q&A
What is Vald?
Vald is a cloud-native approximate nearest neighbor search engine for dense vectors, built to scale horizontally across many machines. It suits engineering teams running similarity search over very large embedding collections on Kubernetes.
Which languages have Vald SDKs?
Official SDKs are available for Go, Java, Node.js and Python, all talking to Vald over gRPC.
How much does Vald cost?
Vald is 100% free.
Is there a free version of Vald?
Yes. Vald is completely free to use.
What are the main features of Vald?
Distributed ANN search, Asynchronous auto-indexing, Index replication, Automatic index backup, Custom gRPC filters, Multi-language SDKs.
Who makes Vald?
Vald is made by vdaas.
What are the best alternatives to Vald?
Popular alternatives include Elasticsearch, Qdrant, Weaviate.