Can you mix embeddings from two different models in one index?
PICTURE THIS: DATABASE INDEX
WHY — Tokens instead of words?
LLMs use tokens (not full words) because it helps them:
who only read docs from people who shipped.
and tied to GenAI / LLM work.
Each piece maps to a number the network can learn.
Fixed pieces are what transformers expect as input.
STEPS — What happens with tokens?
Before the model can read a sentence, it goes through these steps:
- 1Each embedding model defines
its own space and scale.
- 2Vectors from model A
are not comparable to vectors from model B.
- 3Rebuild the index if
you change the embedding model.
- 4Context mix
Attention looks at nearby tokens together.
- 5Next token
The model scores what should come next.
- 6Decode
IDs turn back into readable text.
EXAMPLE — See it in action
Let's see how a real sentence is tokenized (tokens may vary by model):
Note: Actual tokens and IDs depend on the tokenizer (e.g., GPT, Llama, etc.).
Key takeaway
Each embedding model defines its own space and scale. Vectors from model A are not comparable to vectors from model B.