API Documentation

Quick Start

1. 1
   Register — Create your account at /register. Billing starts only after you add funds to your account.
2. 2
   Get your API key — Navigate to Dashboard → API Keys and generate a new key. Keys start with sk-rr-.
3. 3
   Set the base URL — Configure your client to use http://43.224.34.216/v1 as the base URL, then call /models to see the live model list.

OpenRouter-style quick call

The lowest-friction path mirrors OpenRouter: one base URL, one Bearer key, discover models first, then send a chat request. Optional attribution headers are forwarded when an OpenRouter upstream channel is selected.

curl http://43.224.34.216/v1/models \
  -H "Authorization: Bearer sk-rr-your-key-here"

curl http://43.224.34.216/v1/chat/completions \
  -H "Authorization: Bearer sk-rr-your-key-here" \
  -H "Content-Type: application/json" \
  -H "HTTP-Referer: https://your-app.example" \
  -H "X-OpenRouter-Title: Your App" \
  -d '{
    "model": "openai/gpt-5.2",
    "messages": [{"role": "user", "content": "Hello from Haze"}]
  }'

Staging currently verifies hekiu through the Anthropic-native /v1/messages path. OpenRouter model IDs become live after an admin adds an active OpenRouter channel with matching models.

OpenAI SDK — Python

Install the official OpenAI Python library and point it at Haze.

pip install openai

from openai import OpenAI

client = OpenAI(
    api_key="sk-rr-your-key-here",
    base_url="http://43.224.34.216/v1"
)

response = client.chat.completions.create(
    model="openai/gpt-5.2",
    messages=[{"role": "user", "content": "Hello!"}]
)
print(response.choices[0].message.content)

OpenAI SDK — Node.js

Works with the official openai npm package.

npm install openai

import OpenAI from 'openai'

const client = new OpenAI({
  apiKey: 'sk-rr-your-key-here',
  baseURL: 'http://43.224.34.216/v1',
  defaultHeaders: {
    'HTTP-Referer': 'https://your-app.example',
    'X-OpenRouter-Title': 'Your App',
  },
})

const response = await client.chat.completions.create({
  model: 'openai/gpt-5.2',
  messages: [{ role: 'user', content: 'Hello!' }],
})
console.log(response.choices[0].message.content)

Claude Code

Set environment variables before launching Claude Code. This routes all Claude API calls through Haze.

export ANTHROPIC_BASE_URL=http://43.224.34.216
export ANTHROPIC_API_KEY=sk-rr-your-key-here
claude

Or set them inline for a single session: ANTHROPIC_BASE_URL=... ANTHROPIC_API_KEY=... claude

Cursor

Configure Cursor to use Haze as an OpenAI-compatible backend for the supported endpoints on this page.

1. 1Open Cursor → Settings (⌘,)
2. 2Navigate to Models tab
3. 3Under OpenAI API Base, enter: http://43.224.34.216/v1
4. 4Enter your Haze key (starts with sk-rr-) in the API key field
5. 5Click Verify and save

Supported Models

This table shows common examples. The source of truth isGET /v1/modelsfor the active channels in this deployment.

Rate Limits & Errors

Haze returns standard HTTP status codes. Your per-key rate limits are configured in the dashboard.