TL;DR: # Supabase pgvector indexes: pick deliberately, not by copy-paste An unindexed vector column does a sequential scan per query. That is fine for a thousand rows and unusable for a million. Agents either never index or cargo-cult parameters from a blog post written for a different scale. ## Checkable procedure 1.

## Fix
1. # Supabase pgvector indexes: pick deliberately, not by copy-paste An unindexed vector column does a sequential scan per query.
   Expected: this specific failure stops.
2. Start with HNSW.
   Expected: this specific failure stops.
3. A fast index returning the wrong neighbors is worse than a slow correct scan. ## Quick test Run your top 20 production-like queries with.
   Expected: this specific failure stops.
4. Then spot-check that the nearest neighbors match a brute-force scan on a sample; recall problems show up here first.
   Expected: this specific failure stops.

## Details
That is fine for a thousand rows and unusable for a million. Agents either never index or cargo-cult parameters from a blog post written for a different scale. ## Checkable procedure 1. 2. Consider IVFFlat when you have a large static dataset and want to tune the recall/speed tradeoff explicitly with lists and probes. It needs enough rows to build useful lists; on small tables it is worse than no index. 3. Match the index to the distance operator you query with (cosine, L2, inner product). An index built for one operator does not accelerate queries using another. 4. Build the index after loading the initial data, not before. Building on an empty table and then bulk-loading gives a poor structure; reindex after the bulk load. 5. Measure recall on your own queries, not just latency. `explain` and confirm the index is used.

## When to use
You hit exactly this: Supabase pgvector indexes: HNSW for most cases, IVFFlat when you need the tradeoff in Supabase pgvector indexes.

## When not to use
A different error, or the same symptom in a different tool. This page only covers the failure above.

## Compatibility
Supabase pgvector indexes.

## Root cause
It is the better default for most workloads: good recall, no training step, and it handles incremental inserts well.