Embeddings

OpenAI-compatible embeddings endpoint.

POST https://api.ecohash.com/v1/embeddings

Headers

Header	Value
Authorization	Bearer eco_YOUR_KEY
Content-Type	application/json

Request body

Field	Type	Required	Notes
model	string	yes	Embedding model ID
input	string or array of strings	yes	One or many texts to embed. Batch with an array for efficiency
encoding_format	string	no	"float" (default) or "base64"

Example request

curl https://api.ecohash.com/v1/embeddings \
  -H "Authorization: Bearer eco_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "bge-large-en-v1.5",
    "input": ["EcoLink makes GPU compute simple.", "Another sentence to embed."]
  }'

Response body

{
  "object": "list",
  "data": [
    { "object": "embedding", "index": 0, "embedding": [0.012, -0.004, ...] },
    { "object": "embedding", "index": 1, "embedding": [0.019, -0.007, ...] }
  ],
  "model": "bge-large-en-v1.5",
  "usage": {
    "prompt_tokens": 12,
    "total_tokens": 12
  }
}

Embedding dimension depends on the model — typically 768, 1024, or 1536 floats.

Batch efficiency

Embedding 100 strings in one call is several times cheaper (per-token) than 100 separate calls, because overhead dominates small requests. For typical RAG pipelines, batch 32–256 strings per call.

Errors

See Errors. Common: 401, 402, 404, 429.