Hanzo Embeddings

Hanzo Embeddings turns text into vectors through a single OpenAI-compatible endpoint on the AI gateway. Generate embeddings with any supported model, then store and query them in Vector for semantic search and RAG.

Generate Embeddings

The endpoint is drop-in compatible with the OpenAI embeddings API, so existing SDKs work by pointing at api.hanzo.ai.

curl https://api.hanzo.ai/v1/embeddings \
  -H "Authorization: Bearer hk-..." \
  -H "Content-Type: application/json" \
  -d '{
    "model": "text-embedding-3-small",
    "input": "Hanzo is an AI cloud platform."
  }'

Response (OpenAI shape):

{
  "object": "list",
  "data": [{ "object": "embedding", "index": 0, "embedding": [0.021, -0.014, "..."] }],
  "model": "text-embedding-3-small",
  "usage": { "prompt_tokens": 8, "total_tokens": 8 }
}

Pass an array of strings as input to embed a batch in one request.

Store and Search

Provision a Vector collection whose dimensions match your model, then upsert the returned vectors and query for nearest neighbors.

from openai import OpenAI

client = OpenAI(api_key="hk-...", base_url="https://api.hanzo.ai/v1")

vecs = client.embeddings.create(
    model="text-embedding-3-small",
    input=["first document", "second document"],
)
# Upsert vecs.data[i].embedding into your Vector collection, then search.

Models

Model availability and pricing follow the LLM gateway. Because embeddings share the gateway's per-key budgets and rate limits, the same hk- key governs spend across chat and embeddings.

Related

• Vector — store and search the embeddings you generate
• Search — hybrid full-text + semantic ranking
• LLM Gateway — one API for 200+ models, including embeddings
• API Keys — budgets and rate limits per key
• API Reference — every endpoint at api.hanzo.ai