Working Applications

Applications and tools that have been successfully tested and verified to work with Zaguán's OpenAI-compatible gateway. These integrations demonstrate real-world usage and compatibility.

Fully Working

These applications have been thoroughly tested and work seamlessly with Zaguán.

Qwen-CodeVerified

AI-powered code generation and assistance

GitHub

✓
Full compatibility with Zaguán's OpenAI-compatible API
✓
Code completion and generation working as expected
✓
All features tested and operational

Configuration

Set these environment variables to configure Qwen-Code with Zaguán:

export OPENAI_API_KEY="ps_live_YourZaguanAPIKey"
export OPENAI_BASE_URL="https://api.zaguanai.com/v1"
# Backup URL in case of connection issues
# export OPENAI_BASE_URL="https://api-eu-fi-01.zaguanai.com/v1"
export OPENAI_MODEL="zaguanai/minimax-m2"

After setting these variables, run qwen to start coding with MiniMax M2.

Shell Script Wrapper (Recommended)

Create a wrapper script to avoid affecting your global environment. Save this as ~/bin/qcode:

#!/usr/bin/bash
export OPENAI_API_KEY="ps_live_YourZaguanAPIKey"
export OPENAI_BASE_URL="https://api.zaguanai.com/v1"
# Backup URL in case of connection issues
# export OPENAI_BASE_URL="https://api-eu-fi-01.zaguanai.com/v1"
export OPENAI_MODEL="zaguanai/minimax-m2"

qwen $@

unset OPENAI_API_KEY
unset OPENAI_BASE_URL
unset OPENAI_MODEL

Make it executable: chmod 750 ~/bin/qcode

Then run qcode instead of qwen to use Zaguán without modifying your shell environment.

ZedVerified

High-performance code editor with AI assistance

Website

✓
Native OpenAI API support works perfectly with Zaguán
✓
AI-powered code completion and suggestions
✓
Seamless integration with all Zaguán features

Configuration

Edit ~/.config/zed/settings.json and add the following configuration:

{
  "buffer_font_size": 17.0,
  "theme": "Ayu Dark",
  "edit_predictions": {
    "enabled_in_text_threads": true
  },
  "show_edit_predictions": true,
  "agent": {
    "default_model": {
      "provider": "Zaguan",
      "model": "zaguanai/minimax-m2"
    },
    "always_allow_tool_actions": true,
    "model_parameters": []
  },
  "disable_ai": false,
  "language_models": {
    "openai_compatible": {
      "Zaguan": {
        "api_url": "https://api.zaguanai.com/v1",
        "available_models": [
          {
            "name": "zaguanai/kimi-k2-thinking",
            "display_name": "Zaguán Kimi K2 Thinking",
            "max_tokens": 32768,
            "capabilities": {
              "tools": true,
              "images": false,
              "parallel_tool_calls": true,
              "prompt_cache_key": false
            }
          },
          {
            "name": "zaguanai/minimax-m2",
            "display_name": "Zaguán MiniMax M2",
            "max_tokens": 32768,
            "capabilities": {
              "tools": true,
              "images": false,
              "parallel_tool_calls": true,
              "prompt_cache_key": false
            }
          }
        ]
      }
    }
  }
}

This configuration sets up two models: Kimi K2 Thinking and MiniMax M2 (default).

Running Zed

Launch Zed with your Zaguán API key:

ZAGUAN_API_KEY="ps_live_YourZaguanAPIKey" zed

Both configured models will be available in Zed's AI assistant. You can switch between them in the editor.

Partial Support

These applications work with Zaguán but may require additional configuration or have known limitations.

OpenAI CodexPartial

OpenAI's code generation model

GitHub

Working:

✓
Basic connection and authentication
✓
Chat functionality operational

Known Issues:

⚠
Code editing features require additional configuration
⚠
Some advanced features may need manual setup

Status: Basic functionality works, but full feature support requires additional configuration work. Contributions and testing welcome!

Testing Your Application

Want to test if your application works with Zaguán? Here's how to get started:

Configure the Base URL

Point your application's OpenAI API endpoint to https://api.zaguanai.com/v1

Add Your API Key

Use your Zaguán API key instead of an OpenAI key. Get yours from the dashboard.

Choose Your Model

Use any of our 500+ models with the provider/model format.

Test and Share

Test your integration and let us know how it works! We'd love to add your application to this list.

Contribute Your Experience

Have you successfully integrated Zaguán with an application? We'd love to hear about it!

What to Share:

•
Application name and description
•
What features work and what doesn't
•
Any configuration tips or gotchas
•
Setup instructions if applicable

Reach out through our support channels or submit a pull request to help other users discover compatible applications.