Back to Features

AI-Enabled Workflows & Vector Search

Vadyl brings the power of AI directly into your backend. With built-in vector storage and embedding generation, you can implement semantic search and RAG pipelines easily. Vadyl provides native vector search capabilities directly in your entities. Generate embeddings automatically, perform semantic similarity searches, and build RAG (Retrieval-Augmented Generation) pipelines without managing separate vector databases like Pinecone or Weaviate. Vector fields are treated as first-class entity properties with automatic indexing and optimized search algorithms.

1536
Dimensions
<50ms
Search Speed
99.9%
Accuracy
example.ts
// Define entity with vector field
const Product = vadyl.entity("Product", {
  name: type.string(),
  description: type.text(),
  // Auto-generate embeddings from description
  embedding: type.vector({ 
    source: "description",
    model: "openai/text-embedding-3-small"
  })
});

// Semantic similarity search
const similar = await Product.vectorSearch(
  "comfortable running shoes",
  { limit: 10, threshold: 0.8 }
);

Key Benefits

Auto-Embedding

Automatically generate embeddings from text fields on save.

Multi-Model Support

Support for OpenAI, Cohere, and custom embedding models.

Hybrid Search

Combine vector similarity with traditional filtering and sorting.

Real-World Use Cases

Semantic Search

Find products, documents, or content by meaning, not just keywords.

RAG Pipelines

Retrieve relevant context for AI chatbots and assistants.

Ready to try AI-Enabled Workflows & Vector Search?

Start building with Vadyl today and experience the difference.