DEVELOPER SDK
Zero Labs API Quick Start
Start generating real-time completions with Zero Titan models in seconds using JavaScript, Python, or cURL.
import OpenAI from 'openai';
// Initialize OpenAI client pointing to Zero Labs API Gateway
const client = new OpenAI({
baseURL: 'https://www.zerolabs.live/api/v1',
apiKey: process.env.ZERO_API_KEY || 'zl-sk-live-your_api_key_here',
});
async function main() {
// 1. Standard Non-Streaming Call
const response = await client.chat.completions.create({
model: 'titan-pro', // 128k context balanced model (or 'titan-ultra')
messages: [
{ role: 'user', content: 'Explain 128k context reasoning in 3 bullet points.' }
],
});
console.log(response.choices[0].message.content);
// 2. Real-Time Streaming Call
const stream = await client.chat.completions.create({
model: 'titan-pro',
messages: [
{ role: 'user', content: 'Write a TypeScript debounce function.' }
],
stream: true,
});
for await (const chunk of stream) {
process.stdout.write(chunk.choices[0]?.delta?.content || '');
}
}
main();Authentication
The Zero Labs API uses API keys for request authentication. Provide your secret key in the Authorization HTTP header.
Authorization: Bearer YOUR_API_KEY
Chat Completions
POSThttps://www.zerolabs.live/api/v1/chat/completions
Creates a model response for the given conversation messages. Supports both synchronous output and Server-Sent Event (SSE) streaming.
| Parameter | Type | Required | Description |
|---|---|---|---|
| model | string | Yes | "titan-pro" or "titan-ultra" |
| messages | array | Yes | Array of message objects with role and content |
| stream | boolean | Optional | If set, partial message deltas will be sent as SSE chunks. Default: false |
| temperature | number | Optional | Sampling temperature between 0.0 and 2.0. Default: 1.0 |
Available Models
Titan Pro (Thinking)
BALANCEDBalanced performance for everyday reasoning, RAG embeddings, and rapid coding workflows.
Context: 128,000 tokens
Latency: <500ms P50
Titan Ultra (Thinking)
PROFrontier multi-step reasoning, mathematical proof, code synthesis, and deep logic.
Context: 128,000 tokens
Self-Verification CoT Engine
