Embeddings
carpe_embeddingsGenerate OpenAI-compatible embedding vectors for 1–128 texts, selecting TEE-only or router routes to balance privacy and per-token cost.
Instructions
Compute embedding vectors (OpenAI-compatible) for 1–128 texts. route "carpe" (/v1) or "router" (/router, cheapest market; privacy applies). Paid per input token.
Tool call template
Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.
{
"id": 1,
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"arguments": {
"input": "<input>"
},
"name": "carpe_embeddings"
}
}Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| input | Yes | Text or array of up to 128 texts. | |
| model | No | Embedding model id (default: text-embedding-bge-m3). See carpe_list_models type "embedding". | |
| route | No | "carpe" → /v1: Carpe alone, always in its TEE, Carpe price. "router" → /router: best price across markets (may be served outside a TEE; see privacy). Default from CARPE_DEFAULT_ROUTE. | |
| privacy | No | route "router" only. true = TEE markets only (Carpe, Chutes, Venice); "private" = for models Venice runs privately, only markets that keep nothing. Ignored on route "carpe", which is always TEE. | |
| dimensions | No | ||
| encoding_format | No |