What breaks when you put vectors from TWO different embedding models in one index?

Attempt it

Quick reference

TL;DR

Mixing vectors from two embedding models in one index breaks retrieval silently. The two models live in unrelated coordinate systems, so cross-model cosine is noise; no error is raised.

Memory aid

Sign in for the mnemonic that makes this stick. See Pro pricing.

Easy to grasp

Imagine two friends who each invent their own private map of a city. On one friend's map, north is up. On the other's, north is to the right. The street names look similar, but the grid coordinates do not line up. If you ask 'which point is closest to (3, 4)?' the answer depends on whose map you are reading. Library catalog systems work the same way. Two librarians each build a private filing scheme of meaning. Drop addresses from both schemes into one drawer and ask 'which is nearest my new search?' and you get nearest by accident answers, not nearest by meaning, and nothing in the drawer warns you the schemes do not match.

Key concepts

Embeddings Vector databases

Detailed answer & concept explanation

~4 min read

Everything you need to truly understand this topic: intuition, mechanics, step by step explanation, code, formulas, and worked example. Click to expand.

Everything you need to truly understand this topic: intuition, mechanics, step by step explanation, code, formulas, and worked example.

Sign in for 5 free unlocks/day or go Pro for unlimited. See Pro pricing.

Quick scan · 2–4 min

Everything important, quickly.

2 min: same dim != same space + DB silence on mixing + parallel-index migration + model_id metadata discipline + detection via labeled-set drift.

Real-world examples

Real products, models, and research that use this idea.

Pinecone, Weaviate, Qdrant, and pgvector all accept any vector of the indexed dimension; none enforces a model-id check by default. The discipline lives in the application layer.
OpenAI's embedding migration docs explicitly recommend a parallel-index pattern with full re-embedding when moving from text-embedding-3-small to text-embedding-3-large.
Production RAG stacks at Notion and Cursor store model_id and model_version as metadata on every vector so a mixed-model state is at least detectable in audits.

Interviewer follow-ups

What an interviewer would ask next. Try answering before peeking at the approach.

QWhy does the vector database not raise on a model mismatch?

Sign in or go Pro to see the approach for this follow-up.

QHow would you migrate a 500M-doc index to a new embedding model with zero downtime?

Sign in or go Pro to see the approach for this follow-up.

QWhat metric would alert you to silent mixing in production?

Sign in or go Pro to see the approach for this follow-up.

Don't say this

Red flags and common mistakes that signal junior thinking. Click to expand.

Most common mistake

Assuming that two embedding models with the same output dimension are interchangeable, so you can save cost by leaving old vectors alone after a model upgrade.

Sign in or go Pro to see all red flags and common mistakes. See Pro pricing.

60-second cheat sheet

The night-before-the-interview bullets. Scan these on the way to the call.

Sign in for 5 unlocks/day or go Pro for unlimited revision sheets. See Pro pricing.

References

Primary sources. Skim if you want the original framing.