What if your database had built-in full-text search, vector embeddings, file storage, and analytics — without adding Elasticsearch, Pinecone, S3, and a data warehouse?
What is Xata?
Xata is a serverless database platform built on PostgreSQL and Elasticsearch. It combines a relational database with full-text search, vector search, file attachments, and analytics in one API.
Why Xata
1. PostgreSQL + Search in One
import { XataClient } from './xata';
const xata = new XataClient();
// Full-text search (no Elasticsearch setup needed)
const results = await xata.db.posts.search('react server components', {
target: ['title', 'content'],
fuzziness: 1,
highlight: { enabled: true },
});
2. AI / Vector Search
// Ask questions about your data
const answer = await xata.db.docs.ask('How do I configure authentication?', {
searchType: 'vector',
vectorSearch: {
column: 'embedding',
contentColumn: 'content',
},
});
// Returns AI-generated answer based on your data
3. Type-Safe Client (Auto-Generated)
xata codegen
// Auto-generated from your schema
const user = await xata.db.users.create({
name: 'Alice', // string - typed
email: 'a@b.com', // string - typed
age: 30, // number - typed
avatar: fileBuffer, // file attachment - typed
});
// Filtering with type safety
const activeUsers = await xata.db.users
.filter({ age: { $gt: 18 }, 'settings.notifications': true })
.sort('name', 'asc')
.getPaginated({ pagination: { size: 20 } });
4. File Attachments
// Upload file directly to a record
await xata.db.products.update('rec_123', {
image: {
base64Content: imageBase64,
mediaType: 'image/png',
name: 'product.png',
},
});
// Get file URL with transformations
const url = product.image?.url;
const thumbnail = `${url}?tr=w-200,h-200`;
5. Aggregations
const stats = await xata.db.orders.aggregate({
totalRevenue: { sum: { column: 'amount' } },
avgOrderValue: { average: { column: 'amount' } },
ordersByMonth: {
dateHistogram: { column: 'created_at', calendarInterval: 'month' },
},
topProducts: {
topValues: { column: 'product_name', size: 10 },
},
});
6. Branching (Like Git)
# Create a branch for testing
xata branch create feature-new-schema
# Make schema changes on the branch
xata schema edit --branch feature-new-schema
# Merge when ready
xata branch merge feature-new-schema
Free Tier
| Feature | Free |
|---|---|
| Records | 750K |
| Storage | 2GB |
| Search | 250K queries/month |
| AI questions | 250/month |
| File storage | 2GB |
| Branches | 15 |
Xata vs Supabase vs PlanetScale
| Xata | Supabase | PlanetScale | |
|---|---|---|---|
| Database | PostgreSQL | PostgreSQL | MySQL |
| Full-text search | Built-in | Via pg_trgm | No |
| Vector search | Built-in | Via pgvector | No |
| File storage | Built-in | Separate bucket | No |
| Branching | Yes | No | Yes |
| Type generation | Auto | Via PostgREST types | No |
| AI queries | Built-in | Via Edge Functions | No |
Getting Started
npm install @xata.io/cli -g
xata init
xata codegen
The Bottom Line
Xata is the database that does everything. PostgreSQL for your data, Elasticsearch for your search, vector DB for your AI, and file storage for your uploads — all in one API.
Need data tools? I build scraping solutions. Check my Apify actors or email spinov001@gmail.com.
