Endpoint
POST https://api.plugsky.com/v1/embeddings
Request body
json
{
"model": "plugsky-embed-v1",
"input": "Plugsky is an OpenAI-compatible AI platform.",
"encoding_format": "float"
}Parameters
| Param | Type | Notes |
|---|---|---|
model | string | Required. plugsky-embed-v1 (1536d) or plugsky-embed-large (3072d). |
input | string | array | Required. Up to 2,048 strings per request. |
encoding_format | string | float or base64. Default float. |
user | string | Per-end-user identifier. |
dimensions | integer | Optional. Truncate to N dimensions (faster, lower storage). |
Response
json
{
"object": "list",
"data": [
{
"object": "embedding",
"index": 0,
"embedding": [0.023, -0.012, 0.041, ...]
}
],
"model": "plugsky-embed-v1",
"usage": {
"prompt_tokens": 12,
"total_tokens": 12
}
}Frequently asked questions
Are embeddings cached?
No — every request computes a fresh embedding. Cache in your own layer (Redis, your DB) if you re-embed the same content.
What languages are supported?
plugsky-embed-v1 supports 50+ languages with strong cross-lingual retrieval. plugsky-embed-large is best for multilingual and Arabic.
Can I bring my own embedding model?
On Enterprise contracts, yes. We support Cohere, Voyage, BGE, and custom models via the model router.
How do I use these for RAG?
Pair with the RAG API: chunk your documents, embed them, store in a vector DB, then query for retrieval.
Try embeddings
OpenAI-ada-compatible. 1536d or 3072d vectors. Flat monthly pricing.
Start $5 trial → See feature overview