Generate text embeddings using OpenAI and Google models through the embeddings API

Embeddings

osmAPI supports text embedding models through the OpenAI-compatible /v1/embeddings endpoint. Embedding models convert text into numerical vectors that capture semantic meaning, useful for search, similarity, clustering, and RAG applications.

Available Models

Model	Provider	Price / 1M tokens	Dimensions	Max Input
`text-embedding-3-small`	OpenAI	$0.02	1,536	8,191
`text-embedding-3-large`	OpenAI	$0.13	3,072	8,191
`gemini-embedding-001`	Google AI Studio	$0.15	3,072	2,048
`gemini-embedding-2-preview`	Google AI Studio	$0.20	3,072	8,192

You can also view all available embedding models via the dedicated endpoint.

Making Requests

Single Text Input

curl -X POST "https://api.osmapi.com/v1/embeddings" \
  -H "Authorization: Bearer $OSM_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "text-embedding-3-small",
    "input": "The quick brown fox jumps over the lazy dog"
  }'

Multiple Text Inputs (Batch)

curl -X POST "https://api.osmapi.com/v1/embeddings" \
  -H "Authorization: Bearer $OSM_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "text-embedding-3-small",
    "input": [
      "First document to embed",
      "Second document to embed",
      "Third document to embed"
    ]
  }'

Specifying a Provider

Use the provider/model format to route to a specific provider:

curl -X POST "https://api.osmapi.com/v1/embeddings" \
  -H "Authorization: Bearer $OSM_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "google-ai-studio/gemini-embedding-001",
    "input": "Hello world"
  }'

Response Format

The response follows the OpenAI embeddings format:

{
  "id": "embd-abc123",
  "object": "list",
  "data": [
    {
      "object": "embedding",
      "embedding": [0.0023, -0.0091, 0.0152, ...],
      "index": 0
    }
  ],
  "model": "openai/text-embedding-3-small",
  "usage": {
    "prompt_tokens": 9,
    "total_tokens": 9,
    "cost": 0.0000166
  }
}

The usage.cost field shows the total cost in USD for the request.

Parameters

Parameter	Type	Required	Description
`model`	string	Yes	Model ID (e.g., `text-embedding-3-small` or `openai/text-embedding-3-small`)
`input`	string, string[], number[], or number[][]	Yes	Text or token array to embed. Can be a single string, array of strings, token array, or array of token arrays.
`dimensions`	integer	No	Number of output dimensions. Supported by OpenAI v3 models and Google models.
`encoding_format`	string	No	Format of the embedding vectors. `float` (default) or `base64`.

Custom Dimensions

Both OpenAI v3 models and Google models support reducing the output dimensions:

curl -X POST "https://api.osmapi.com/v1/embeddings" \
  -H "Authorization: Bearer $OSM_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "text-embedding-3-large",
    "input": "Hello world",
    "dimensions": 1024
  }'

Reducing dimensions trades some accuracy for smaller storage and faster similarity comparisons. OpenAI's text-embedding-3-large at 1024 dimensions still outperforms the full 1536-dimension ada-002 model.

List Embedding Models

You can list all available embedding models:

curl "https://api.osmapi.com/v1/embeddings/models" \
  -H "Authorization: Bearer $OSM_API_KEY"

Pricing

Embeddings are billed on input tokens only (no output cost). The cost is deducted from your organization's credits when using credits or hybrid mode, or charged to your provider API key in API keys mode.

SDK Compatibility

The endpoint is fully compatible with the OpenAI SDK:

from openai import OpenAI

client = OpenAI(
    api_key="your-osm-api-key",
    base_url="https://api.osmapi.com/v1"
)

response = client.embeddings.create(
    model="text-embedding-3-small",
    input="Hello world"
)

print(response.data[0].embedding[:5])

import OpenAI from "openai";

const client = new OpenAI({
	apiKey: "your-osm-api-key",
	baseURL: "https://api.osmapi.com/v1",
});

const response = await client.embeddings.create({
	model: "text-embedding-3-small",
	input: "Hello world",
});

console.log(response.data[0].embedding.slice(0, 5));

Embeddings

On this page