Skip to main content
POST
/
v1
/
embeddings
Embeddings
curl --request POST \
  --url https://modelswitch.io/v1/embeddings \
  --header 'Content-Type: application/json' \
  --data '
{
  "model": "<string>",
  "input": {},
  "encoding_format": "<string>"
}
'
{
  "object": "<string>",
  "data": [
    {
      "object": "<string>",
      "index": 123,
      "embedding": [
        {}
      ]
    }
  ],
  "model": "<string>",
  "usage": {
    "prompt_tokens": 123,
    "total_tokens": 123
  }
}
Convert text into a vector representation (embedding). Embeddings are used for semantic search, clustering, similarity comparisons, and retrieval-augmented generation (RAG). Authentication: Authorization: Bearer ms-YOUR_KEY

​
Parameters

​
model
string
required
Embedding model ID. See available models below. Example: text-embedding-3-small.
​
input
string or array
required
Text to embed. Pass a single string for one embedding, or an array of strings to generate multiple embeddings in a single request.
​
encoding_format
string
default:"float"
Format for the returned embedding vector. Either "float" (array of numbers) or "base64" (base64-encoded string).

​
Examples

​
Single input

curl https://modelswitch.io/v1/embeddings \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer ms-YOUR_KEY" \
  -d '{
    "model": "text-embedding-3-small",
    "input": "The quick brown fox",
    "encoding_format": "float"
  }'
Request body: 
{
  "model": "text-embedding-3-small",
  "input": "The quick brown fox",
  "encoding_format": "float"
}
Response: 
{
  "object": "list",
  "data": [{
    "object": "embedding",
    "index": 0,
    "embedding": [0.0023, -0.0091, 0.0156, "... 1536 dimensions"]
  }],
  "model": "text-embedding-3-small",
  "usage": {
    "prompt_tokens": 5,
    "total_tokens": 5
  }
}

​
Batch input

Pass an array of strings to embed multiple texts in one request. The response data array preserves the input order via the index field. 
{
  "model": "text-embedding-3-small",
  "input": ["First sentence", "Second sentence", "Third sentence"]
}
Batching multiple texts in a single request is more efficient than making separate calls. Each string in the array is embedded independently; they do not affect each other.

​
Response

​
object
string
Always "list".
​
data
array
Array of embedding objects, one per input string.
​
model
string
The model used to generate the embeddings.
​
usage
object
Token counts for billing purposes.

​
Available models

ModelDimensionsNotes
text-embedding-3-small1536Cost-effective, recommended for most use cases
text-embedding-3-large3072Higher quality, better for precision-sensitive tasks
text-embedding-ada-0021536Legacy model
Use GET /v1/models for the full, up-to-date list of available embedding models.