Embeddings

POST /v1/embeddings

Generate a vector embedding (or batch of embeddings) for one or more text inputs. Compatible with the OpenAI Embeddings API.

Headers

Request body

Examples

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

curl -X POST https://api.quantized.us/v1/embeddings \
  -H "Authorization: Bearer sk-quantized-YOUR-KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "text-embedding-3-small",
    "input": ["one", "two", "three"]
  }'

from openai import OpenAI

client = OpenAI(
    api_key="sk-quantized-YOUR-KEY",
    base_url="https://api.quantized.us/v1",
)

resp = client.embeddings.create(
    model="text-embedding-3-small",
    input=["one", "two", "three"],
)
for item in resp.data:
    print(item.index, len(item.embedding))

import httpx

response = httpx.post(
    "https://api.quantized.us/v1/embeddings",
    headers={"Authorization": "Bearer sk-quantized-YOUR-KEY"},
    json={
        "model": "text-embedding-3-small",
        "input": "Hello world",
        "dimensions": 512,
    },
)
data = response.json()
print(len(data["data"][0]["embedding"]))  # 512

Response

{
  "object": "list",
  "model": "text-embedding-3-small",
  "data": [
    {
      "object": "embedding",
      "index": 0,
      "embedding": [0.0123, -0.0456, 0.0789, ...]
    }
  ],
  "usage": {
    "prompt_tokens": 10,
    "total_tokens": 10,
    "credits_used": 32,
    "credits_remaining": 999968
  }
}

Response fields

Models

When routing through OpenRouter, prefix the model with openai/ (e.g. openai/text-embedding-3-small). The same rate table is applied via fallback because OpenRouter does not include cost data in embedding responses.

Discovery via /v1/models

The three OpenAI embedding models are listed in GET /v1/models alongside chat models. Filter for embeddings either by:

• supported_features contains "embeddings" (recommended, semantic):

  embedding_models = [m for m in models if "embeddings" in m.get("supported_features", [])]
  

• output_modality.text == false — embedding models declare no media-modality output because the response is a vector, not generated text. Chat models have output_modality.text == true.

The catalog entries also expose pricing in cost.prompt (micro-credits per token).

Providers

# Force OpenRouter passthrough
curl -X POST https://api.quantized.us/v1/embeddings \
  -H "Authorization: Bearer sk-quantized-YOUR-KEY" \
  -H "X-Quantized-Provider: openrouter" \
  -H "Content-Type: application/json" \
  -d '{"model": "openai/text-embedding-3-small", "input": "hello"}'

Errors