Skip to content
QuantCDN

Generate text embeddings for semantic search and RAG applications

POST
/api/v3/organizations/{organisation}/ai/embeddings
curl --request POST \
  --url https://dashboard.quantcdn.io/api/v3/organizations/example/ai/embeddings \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '{ "input": "The Australian government announced new climate policy", "modelId": "amazon.titan-embed-text-v2:0", "dimensions": 1024, "normalize": true }'

Generates vector embeddings for text content using embedding models. Used for semantic search, document similarity, and RAG applications. * * Features: * - Single text or batch processing (up to 100 texts) * - Configurable dimensions (256, 512, 1024, 8192 for Titan v2) * - Optional normalization to unit length * - Usage tracking for billing * * Use Cases: * - Semantic search across documents * - Similarity matching for content recommendations * - RAG (Retrieval-Augmented Generation) pipelines * - Clustering and classification * * Available Embedding Models: * - amazon.titan-embed-text-v2:0 (default, supports 256-8192 dimensions) * - amazon.titan-embed-text-v1:0 (1536 dimensions fixed)

Authorizations

Section titled “Authorizations”

Parameters

Section titled “Parameters”

Path Parameters

Section titled “Path Parameters”
organisation
required
string

The organisation ID

Request Bodyrequired

Section titled “Request Bodyrequired”

Embedding request with single or multiple texts

Media typeapplication/json
object
input
required
One of:
string
modelId

Embedding model to use

string
default: amazon.titan-embed-text-v2:0
dimensions

Output embedding dimensions. Titan v2 supports: 256, 512, 1024, 8192

integer
default: 1024
Allowed values: 256 512 1024 8192
normalize

Normalize embeddings to unit length (magnitude = 1.0)

boolean
default: true
Example
{
  "input": "The Australian government announced new climate policy",
  "modelId": "amazon.titan-embed-text-v2:0",
  "dimensions": 1024,
  "normalize": true
}

Responses

Section titled “Responses”

200

Section titled “200”

Embeddings generated successfully

Media typeapplication/json
object
embeddings
required
One of:

Single embedding vector if input was a string

Array<number>
model
required

Model used to generate embeddings

string
dimension
required

Dimensionality of each embedding vector

integer
usage
required
object
inputTokens
required

Number of tokens in input text(s)

integer
totalTokens
required

Total tokens (same as inputTokens for embeddings)

integer
Example
{
  "embeddings": [
    0.0215,
    0.0008,
    0.0312,
    -0.0087,
    0.0273
  ],
  "model": "amazon.titan-embed-text-v2:0",
  "dimension": 1024,
  "usage": {
    "inputTokens": 8,
    "totalTokens": 8
  }
}

400

Section titled “400”

Invalid request parameters

403

Section titled “403”

Access denied

500

Section titled “500”

Failed to generate embeddings