Quickstart

This path gives you a working integration in minutes, then immediately covers the first reliability checks you should add.

Start from a server environment where you can keep secrets private. Generate an API key in /dashboard/keys, save it in your secret manager, and expose it to runtime as SYLICA_API_KEY. Avoid embedding root keys in browser bundles.

Create and store an API key with least privilege for the environment.
Set base URL to https://api.sylicaai.com/v1 in your OpenAI-compatible SDK.
Send a streaming chat completion and confirm first token latency in logs.
Log x-sylica-request-id for every request to simplify support debugging.

Quick Verification Request

bash

curl https://api.sylicaai.com/v1/chat/completions \
  -H "Authorization: Bearer $SYLICA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "sylica/auto",
    "messages": [
      {"role":"user","content":"Write a two-line haiku about routing."}
    ],
    "stream": true
  }'

TypeScript Example

typescript

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.sylicaai.com/v1",
  apiKey: process.env.SYLICA_API_KEY,
});

const stream = await client.chat.completions.create({
  model: "anthropic/claude-sonnet-4.5",
  messages: [{ role: "user", content: "Hello from Sylica" }],
  stream: true,
});

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

Operational Guidance

As soon as your first call works, add timeouts and retry policy before expanding feature use. Early reliability work prevents hard-to-debug production incidents when request volume grows.