Create embeddings

Embeds up to 100 texts with the workspace embedding model and returns one vector per input, in order. Use it to build your own search or clustering over content that lives outside Dhanu. The call is billed by input tokens; a test key returns deterministic 8-dimensional fixtures at no cost.

Request

curl -X POST https://api.dhanu.ai/api/pub/v1/embeddings \
  -H "Authorization: Bearer $DHANU_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "input": ["momo recipe", "thukpa recipe"] }'

const { embeddings, model } = await dhanu.embeddings.create({
  input: ["momo recipe", "thukpa recipe"],
});

result = dhanu.embeddings.create(["momo recipe", "thukpa recipe"])
vectors = result.embeddings

Response

201 Created

{ "embeddings": [[0.0123, -0.0456, …], [0.0087, -0.0391, …]], "model": "…", "usage": { "inputTokens": 9 }, "cost": { "amount": 1, "currency": "NPR" } }

Errors