Quickstart

Prerequisites

Before you begin, make sure you have the following:

A Unii account and API key (see Authentication for details)
Node.js 18+ for the TypeScript SDK, or Python 3.9+ for the Python SDK

Note

Check your runtime version with node --version or python --version. Older versions may not support the latest SDK features.

Get your API key

You need an API key to authenticate requests. Generate one from the Unii Developer Console by signing in, navigating to Settings, and creating a new key.

Store the key in an environment variable so you don't accidentally commit it to version control:

.env.local

UNII_API_KEY=sk_unii_your_api_key_here

.env.local

UNII_API_KEY=sk_unii_your_api_key_here

Warning

Never hardcode API keys in your source code. Use environment variables or a secrets manager to keep credentials secure.

Install the SDK

Install the Unii SDK using your preferred package manager:

npm install @unii/sdk

npm install @unii/sdk

Tip

TypeScript types are included with the SDK — no additional @types package is needed.

Make your first request

Send a chat completion request to the Unii API. The examples below use the illustrative endpoint POST /v1/chat/completions.

example.ts

import { Unii } from "@unii/sdk";

const client = new Unii({
apiKey: process.env.UNII_API_KEY,
});

const response = await client.chat.completions.create({
model: "unii-latest",
messages: [{ role: "user", content: "Explain RAG in one sentence." }],
});

console.log(response.choices[0].message.content);

example.ts

import { Unii } from "@unii/sdk";const client = new Unii({apiKey: process.env.UNII_API_KEY,});const response = await client.chat.completions.create({model: "unii-latest",messages: [{ role: "user", content: "Explain RAG in one sentence." }],});console.log(response.choices[0].message.content);

Sample response

A successful response returns a JSON object with the generated text:

Response

{
"id": "chatcmpl-abc123",
"object": "chat.completion",
"model": "unii-latest",
"choices": [
  {
    "index": 0,
    "message": {
      "role": "assistant",
      "content": "RAG (Retrieval-Augmented Generation) combines a knowledge base lookup with an LLM to produce answers grounded in your own documents."
    },
    "finish_reason": "stop"
  }
],
"usage": {
  "prompt_tokens": 15,
  "completion_tokens": 28,
  "total_tokens": 43
}
}

Response

{"id": "chatcmpl-abc123","object": "chat.completion","model": "unii-latest","choices": [  {    "index": 0,    "message": {      "role": "assistant",      "content": "RAG (Retrieval-Augmented Generation) combines a knowledge base lookup with an LLM to produce answers grounded in your own documents."    },    "finish_reason": "stop"  }],"usage": {  "prompt_tokens": 15,  "completion_tokens": 28,  "total_tokens": 43}}

Troubleshooting

If your request fails, check the following common issues:

401 Unauthorized — Your API key is missing, expired, or invalid. Verify that UNII_API_KEY is set correctly and try generating a new key from the console.
429 Too Many Requests — You've exceeded the rate limit. Wait a moment and retry, or contact support to discuss higher limits.
Invalid model — The model name you specified doesn't exist. Use unii-latest or check the API Reference for available models.

Important

The API examples on this page are illustrative. Verify the actual request format, response structure, and available models against the latest API Reference before using them in production.

Next steps

Learn how to authenticate your API requests in detail.
Read the API Reference for complete endpoint documentation.
Explore knowledge bases to ground responses in your own data.
Try streaming for real-time responses.

Quickstart

PrerequisitesCopy Anchor Link

Get your API keyCopy Anchor Link

Install the SDKCopy Anchor Link

Make your first requestCopy Anchor Link

Sample responseCopy Anchor Link

TroubleshootingCopy Anchor Link

Next stepsCopy Anchor Link