Skip to main content
Rei Unit provides an OpenAI-compatible API for tools that support custom OpenAI endpoints. Configure the client with the Rei Unit base URL, API key, and model identifier.

Compatible Tools

Clients with configurable OpenAI endpoints include:
  • Cursor
  • OpenAI Codex CLI
  • Continue (VS Code)
  • Aider
  • Cline (VS Code)
  • Open WebUI
  • LM Studio

Connection Details

Use these connection settings: Credit consumption varies by model. The documented options range from rei-qwen3-coder with the lowest consumption to rei-coder-max with the highest. Refer to OpenRouter for current pricing.

Setup Instructions

Cursor IDE

  1. Open Settings → Models → API Keys
  2. Find OpenAI API Key section
  3. Click Override OpenAI Base URL
  4. Enter: https://coder.reilabs.org/v1
  5. Enter your Rei API key
  6. Add model: rei-qwen3-coder

OpenAI Codex CLI

  1. Create config file:
  1. Add configuration:
  1. Set your API key:
  1. Run Codex:

Continue (VS Code)

  1. Install Continue extension
  2. Open Continue settings (~/.continue/config.json)
  3. Add to models array:

Aider

Or set environment variables:

Cline (VS Code)

  1. Install Cline extension
  2. Open Cline settings
  3. Select OpenAI Compatible as provider
  4. Enter Base URL: https://coder.reilabs.org/v1
  5. Enter API Key
  6. Enter Model: rei-qwen3-coder

Open WebUI

  1. Go to Settings → Connections
  2. Add new OpenAI connection
  3. URL: https://coder.reilabs.org/v1
  4. API Key: Your Rei key
  5. The model will auto-discover

Troubleshooting

Test the API directly

You should get a JSON response with the model’s reply.

Common Issues

  • “Invalid API key”: Check your key is correct, no extra spaces
  • “Model not found”: Check that the model identifier matches one of the available models.
  • Connection timeout: Check your internet connection