Getting started

Quickstart

Get from zero to your first proxied AI call in under five minutes. You will change two lines of code — nothing else in your app needs to change.

Steps

1. 1

   Connect a provider API key (admin)

   Go to API Keys → Provider keys → Add key. Paste your OpenAI, Anthropic, or Google API key. Zelyx validates the format and stores it encrypted — your real key never leaves the vault.

   Supported providers: OpenAI, Anthropic, Google, Groq, Together, Mistral, DeepSeek, xAI, OpenRouter, Fireworks, Perplexity, or any custom slug.

2. 2

   Create your Zelyx key

   On the same API Keys page, under Zelyx proxy keys, click New proxy key. Give it a label (e.g. dev-laptop), set an optional daily budget, and click Generate key.

   Your key is shown once. Copy it now — it cannot be retrieved again. It looks like:

   nk_a1b2c3d4e5f6_aBcDeFgHiJkLmNoPqRsTuVwXyZ0123456789abc

3. 3

   Point your SDK at Zelyx

   Change two things in your code: api_key (use your Zelyx key) and base_url (point to Zelyx). Everything else — model names, streaming, tool use, message format — stays exactly the same.

   OpenAI SDK (Python)

   python

   from openai import OpenAI
   
   client = OpenAI(
       api_key="nk_<your-zelyx-key>",
       base_url="https://api.zelyx.app/proxy/openai/v1",
   )
   
   response = client.chat.completions.create(
       model="gpt-4o-mini",
       messages=[{"role": "user", "content": "Hello"}],
   )
   print(response.choices[0].message.content)

   Anthropic SDK (Python)

   python

   import anthropic
   
   client = anthropic.Anthropic(
       api_key="nk_<your-zelyx-key>",
       base_url="https://api.zelyx.app/proxy/anthropic",
   )
   
   message = client.messages.create(
       model="claude-haiku-4-5",
       max_tokens=256,
       messages=[{"role": "user", "content": "Hello"}],
   )
   print(message.content[0].text)

   OpenAI SDK (TypeScript)

   typescript

   import OpenAI from "openai";
   
   const client = new OpenAI({
     apiKey: "nk_<your-zelyx-key>",
     baseURL: "https://api.zelyx.app/proxy/openai/v1",
   });
   
   const response = await client.chat.completions.create({
     model: "gpt-4o-mini",
     messages: [{ role: "user", content: "Hello" }],
   });
   console.log(response.choices[0].message.content);

4. 4

   Verify in the dashboard

   Make a call. Within a few seconds it should appear in Overview and Explorer. Cost, tokens, latency, and model are all recorded automatically.

What happens on each call

When your app sends a request through Zelyx, the proxy:

1. Verifies your Zelyx key and resolves your workspace
2. Decrypts your provider API key from the encrypted vault
3. Runs the gate — checks budgets, model policies, and payment patterns
4. Forwards the request to the provider if approved
5. Streams the response back to your app (recording tokens and latency)
6. Records the cost event in the background for analytics

Your app never sees the provider key. If a budget is exceeded, Zelyx returns HTTP 402 before the provider is contacted — so no tokens are consumed and no cost is incurred.