Building RAG with Supabase pgvector

Supabase pgvector enables vector similarity search directly in PostgreSQL — perfect for RAG applications.

Enable pgvector

```sql

CREATE EXTENSION IF NOT EXISTS vector;

```

Create Embeddings Table

```sql

CREATE TABLE documents (

id BIGSERIAL PRIMARY KEY,

content TEXT NOT NULL,

embedding VECTOR(1536) -- OpenAI ada-002 dimensions

);

CREATE INDEX ON documents USING ivfflat (embedding vector_cosine_ops)

WITH (lists = 100);

```

Store Embeddings

```typescript

import { createClient } from '@supabase/supabase-js'

import OpenAI from 'openai'

const openai = new OpenAI()

const supabase = createClient(URL, KEY)

// Generate embedding

const { data } = await openai.embeddings.create({

model: 'text-embedding-ada-002',

input: 'Your document text here',

})

// Store in Supabase

await supabase.from('documents').insert({

content: 'Your document text here',

embedding: data[0].embedding,

})

```

Query Similar Documents

```sql

CREATE FUNCTION match_documents(query_embedding VECTOR(1536), match_count INT)

RETURNS TABLE (id BIGINT, content TEXT, similarity FLOAT)

AS $$

SELECT id, content, 1 - (embedding <=> query_embedding) AS similarity

FROM documents

ORDER BY embedding <=> query_embedding

LIMIT match_count;

$$ LANGUAGE sql;

```