Embeddings

Use POST /v1/embeddings to create vectors for semantic search, retrieval augmented generation, clustering, and reranking pipelines.

Request

{
  "model": "mistral-embed",
  "input": "ozeye keeps prompts out of storage"
}

Response

{
  "object": "list",
  "data": [{
    "object": "embedding",
    "index": 0,
    "embedding": [0.0023, -0.0192, ...]
  }],
  "model": "mistral-embed",
  "usage": {"prompt_tokens": 8, "total_tokens": 8}
}

The input field may be a string or an array of strings. Only models tagged as embeddings are accepted. If a provider omits usage metadata, ozeye estimates token usage from input bytes for billing.