ZYRAXON Provider Overview

Ten branded models behind one OpenAI-compatible endpoint. What it is, what it costs, and what it will not do.

The ZYRAXON Provider is our own model line, served over an API that speaks the OpenAI chat-completions dialect. You point any OpenAI client at one base URL, choose a ZYRAXON model, and build. There is no provider key to buy, no per-request billing, and no upstream vendor name anywhere in a response.

The whole integration is two URLs

Quickstart with cURL

curl https://zyraxonai.lovable.app/api/public/zyraxon/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{
    "model": "zyraxon-apex",
    "messages": [{ "role": "user", "content": "Hello" }],
    "stream": true
  }'

Quickstart with the OpenAI SDK

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://zyraxonai.lovable.app/api/public/zyraxon/v1",
  apiKey: "not-needed",
});

const stream = await client.chat.completions.create({
  model: "zyraxon-codex",
  messages: [{ role: "user", content: "Build a landing page" }],
  stream: true,
});

for await (const chunk of stream) {
  process.stdout.write(chunk.choices[0]?.delta?.content ?? "");
}

What every response tells you

Read X-ZYRAXON-Input-Modalities when a media answer looks thin: if it lists image but you sent audio too, the audio did not reach the model, and that is a fact you can assert rather than guess.

Connecting from your own tools

The endpoint is all you need, but the wiring differs by client. The Use It From Anywhere guide covers Termux on Android, OpenHands, editors and CLI agents, mobile apps, the SDKs, and workflow runners — each with the exact fields to enter or code to paste.