---
title: "Quickstart"
description: "From API key to first streamed token"
canonical: "https://inference.boundless.network/docs/quickstart"
last-updated: "2018-10-20T01:46:40.000Z"
---

# Quickstart

From API key to first streamed token.


## Try it without an account

Any agent or script can mint an anonymous trial key. The key carries $0.01 of credit, expires after 24 hours, and is tightly rate-limited for demo purposes. Never use it for a real workload.

```bash
curl -X POST https://inference.boundless.network/api/agent-key
```

This is the **agent key**: a free, anonymous API key with no sign-up, no OAuth and no credential. The response carries the key itself, the base URL to point an SDK at, and its limits. The same key can be minted by the `create_agent_key` tool on the [MCP server](/docs/coding-agents-and-harnesses), and a `GET` on the same URL describes the offer without minting anything. It cannot be topped up or extended.

## 1. Get an API key

[Sign up or sign in to the console](/sign-in). While the service is invite-only, a human reviews and activates each new account. Once active, [create an API key yourself](/api-keys); no sales call or support ticket is needed for key creation. Your credit balance is shared across all keys you create.

Customer keys are restricted to inference and served-model discovery. They cannot call account, key, team, billing, or gateway-management routes. The machine-readable grant is published in the [OpenAPI specification](/openapi.json).

## 2. Make your first request

```bash [cURL]
curl https://api.inference.boundless.network/v1/chat/completions \
  -H "Authorization: Bearer $BOUNDLESS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "glm-5.2",
    "messages": [{
      "content": "Say hello",
      "role": "user"
    }]
  }'
```

```python [Python]
import os
from openai import OpenAI

client = OpenAI(
  api_key=os.environ["BOUNDLESS_API_KEY"],
  base_url="https://api.inference.boundless.network/v1",
)

response = client.chat.completions.create(
  model="glm-5.2",
  messages=[{
    "content": "Say hello",
    "role": "user"
  }],
)
```

```typescript [TypeScript]
import OpenAI from 'openai'

const openai = new OpenAI({
  apiKey: process.env.BOUNDLESS_API_KEY,
  baseURL: "https://api.inference.boundless.network/v1",
})

const response = await openai.chat.completions.create({
  model: "glm-5.2",
  messages: [{
    "content": "Say hello",
    "role": "user"
  }],
})
```

## 3. Check your usage

Every response includes a `usage` object with exact token counts, and [the usage page](/usage) shows per-request costs against your credit balance.