Calling an AI API in Node.js
From zero to a working call in JavaScript — LLMs, images and video — using one key so you can switch models without new SDKs. Get a key first: how to get an API key.
Setup
Node 18+ has fetch built in. Keep your key in an env var:
export LINKMODEL_API_KEY="sk_live_..."Your First LLM Call
const res = await fetch("https://api.linkmodel.ai/v1/chat/completions", {
method: "POST",
headers: {
"Authorization": `Bearer ${process.env.LINKMODEL_API_KEY}`,
"Content-Type": "application/json",
},
body: JSON.stringify({
model: "claude-sonnet-4-6",
messages: [{ role: "user", content: "Explain event loops in 2 sentences." }],
}),
});
if (!res.ok) throw new Error(`HTTP ${res.status}`);
console.log(await res.json());Swap model to gpt-5.6-terra, gemini-3.5-flash, or deepseek-v4-flash — one key, no new client. See multimodal AI API.
Generating an Image (Async)
const r = await fetch("https://api.linkmodel.ai/api/v1/image-generation", {
method: "POST",
headers: { "Authorization": `Bearer ${process.env.LINKMODEL_API_KEY}`, "Content-Type": "application/json" },
body: JSON.stringify({ model: "seedream-5.0-pro", prompt: "Hero banner, clean type, 2K" }),
}).then(x => x.json());
const taskId = r.data.task_id; // then poll GET /api/v1/query/image-generation?task_id=Poll GET /api/v1/query/image-generation?task_id=... for images or GET /v1/videos/generations/{task_id} for videos until the task reports success, then read the result URL.
Never Ship Keys to the Browser
Call the AI API from your backend / serverless function, not client-side — a key in frontend JS is trivially stolen. For apps, proxy through your server (same idea as AI API for mobile apps).
Retries with Backoff
async function call(body, tries = 4) {
for (let i = 0; i < tries; i++) {
const r = await fetch(URL, { method: "POST", headers: H, body: JSON.stringify(body) });
if (r.status !== 429) return r.json();
await new Promise(s => setTimeout(s, 2 ** i * 1000)); // exponential backoff
}
throw new Error("rate limited");
}See AI API rate limits.
Bottom Line
fetch, one key, and a model string gets you LLMs, images and video from Node — call it server-side, add backoff, and stream long outputs (streaming guide). Confirm endpoints in the docs.
Start free with a $1 credit and run your first call.
