Embed API

The Embed API allows you to generate vector embeddings from text using multiple AI providers. These embeddings can be used for semantic search, similarity comparison, and other vector-based operations.

API Reference


{
  "description": "API to generate embeddings from text using multiple providers",
  "documentation": "https://apis.do/docs/embed",
  "endpoints": {
    "post": {
      "description": "Generate embeddings from text",
      "parameters": {
        "text": "String: Single text input to embed (alternative to texts)",
        "texts": "Array: Multiple text inputs to embed (alternative to text)",
        "model": "String: Embedding model to use (e.g., \"text-embedding-3-small\")"
      },
      "examples": {
        "singleText": {
          "text": "Embed this text for semantic search",
          "model": "text-embedding-3-small"
        },
        "multipleTexts": {
          "texts": ["Embed this text for semantic search", "And this one too"],
          "model": "text-embedding-3-small"
        }
      }
    }
  }
}

Usage

Single Text Embedding


const response = await fetch('https://apis.do/embed', {
  method: 'POST',
  headers: {
    'Content-Type': 'application/json',
    'Authorization': 'Bearer YOUR_API_KEY'
  },
  body: JSON.stringify({
    text: 'Embed this text for semantic search',
    model: 'text-embedding-3-small'
  })
})
 
const data = await response.json()
// data.data.embeddings contains the embedding vector

Multiple Text Embeddings


const response = await fetch('https://apis.do/embed', {
  method: 'POST',
  headers: {
    'Content-Type': 'application/json',
    'Authorization': 'Bearer YOUR_API_KEY'
  },
  body: JSON.stringify({
    texts: ['Embed this text for semantic search', 'And this one too'],
    model: 'text-embedding-3-small'
  })
})
 
const data = await response.json()
// data.data.embeddings contains an array of embedding vectors

Response Format


// Single text embedding response
{
  "data": {
    "embeddings": [[0.1, 0.2, 0.3, ...]],
    "model": "text-embedding-3-small"
  },
  "usage": {
    "tokens": 5
  }
}
 
// Multiple text embeddings response
{
  "data": {
    "embeddings": [
      [0.1, 0.2, 0.3, ...],
      [0.4, 0.5, 0.6, ...]
    ],
    "model": "text-embedding-3-small"
  },
  "usage": {
    "tokens": 10
  }
}

SDK Usage

You can also use the embedding API through the AI SDK:


import { llm } from 'llm.do'
import { generateEmbeddings } from 'ai'
 
// Generate embeddings
const embeddings = await generateEmbeddings({
  model: llm.embed('text-embedding-3-small'),
  input: ['Embed this text for semantic search', 'And this one too']
})

Supported Models

The embedding API supports various models from different providers:

OpenAI: text-embedding-3-small, text-embedding-3-large
Other providers as configured in the AI providers package

Error Handling

The API returns appropriate error codes and messages:

400: Missing required parameters (text/texts or model)
401: Unauthorized (missing or invalid API key)
500: Server error during embedding generation