Voyage Code 4 is Voyage AI's code-retrieval embedding model, trained so natural-language queries such as bug reports find the right code for coding agents. it succeeds voyage-code-3 at a third lower price.
pick it if
you index source code for coding agents or natural-language code search
you want Voyage's reported 27.54% gain over voyage-code-3 on agentic retrieval
you need 256-2048 Matryoshka dimensions to trade index size for accuracy
skip it if
your corpus is general prose; voyage-4-lite costs a sixth as much
you need open weights to embed code on your own hardware
Voyage 4 Large is the top tier of Voyage AI's Voyage 4 embedding series, a mixture-of-experts model for general-purpose and multilingual retrieval. it sits above voyage-4 and voyage-4-lite, and all three share one embedding space.
pick it if
retrieval quality matters more than cost per token
you want to embed documents with it and run queries on voyage-4-lite
you need multilingual retrieval or 256-2048 dimension embeddings
skip it if
voyage-4 or voyage-4-lite is accurate enough at a lower price
you are indexing source code; voyage-code-4 targets code retrieval
faq
which is cheaper, Voyage Code 4 or Voyage 4 Large?
Voyage Code 4 and Voyage 4 Large cost the same, $0.12/M tokens.
which handles longer documents?
both take up to 32k tokens per text.
can i switch from Voyage Code 4 to Voyage 4 Large without re-indexing?
no. vectors from different embedding models are not comparable, so a switch means re-embedding the whole corpus.
try either model
switch between models inside the same octomind session.