How to implement AI vector search in Laravel 12 with PostgreSQL pgvector embeddings?

Laravel 12 supports pgvector HNSW indexes for semantic search on OpenAI embeddings. Distance operators <-> (cosine), <#> (negative inner product) enable 100ms searches on 1M vectors. Automatic index maintenance during model updates. Powers recommendation systems, chat search in SaaS apps.

Example:-

Code

// Migration
DB::statement('CREATE EXTENSION vector');
DB::statement('ALTER TABLE documents ADD COLUMN embedding vector(1536)');

// Search
Document::whereRaw('embedding <=> ?', [$queryEmbedding])
    ->orderByRaw('embedding <=> ?', [$queryEmbedding])
    ->limit(10)
    ->get();