Code retrieval for coding agents has a new option on OpenRouter.
Voyage Code 4 from @VoyageAI by MongoDB supports Matryoshka embeddings at 256, 512, 1024, and 2048 dimensions, plus multiple quantization options.
Try it: https://t.co/Ac5EoU8XtU
embedding — A list of numbers representing a piece of text's meaning, so that similar meanings end up numerically close and can be searched.
quantization — Shrinking a model by storing its numbers less precisely — like rounding — so it runs faster and fits on smaller hardware, at a small quality cost.
Why it matters
Code retrieval indexes can be tuned for size against recall without re-embeddingA list of numbers representing a piece of text's meaning, so that similar meanings end up numerically close and can be searched.Full definition →, since one Matryoshka model serves 256, 512, 1024 and 2048 dimensions plus several quantizations.
