Warden Integration

AVELIN API is compatible with Warden — Native Swift macOS app supporting multiple AI providers via BYOK. Warden works seamlessly with AVELIN's OpenAI-compatible API endpoint.

Why AVELIN + Warden? Access AVELIN's frontier models with 1M-token context windows, automatic failover, and competitive pricing through a polished desktop interface.

Warden is a native Swift macOS app built on the BYOK (Bring Your Own Key) model. It's written entirely in Swift — no Electron, no web views — which means it launches instantly, uses minimal memory, and feels like a first-party Apple app. The BYOK architecture means Warden never proxies your requests through a middleman server: your API keys stay on your machine, and the only network calls go directly to the AI provider you've configured.

The killer feature is the combination of native Swift performance and zero-trust privacy design. There's no Warden cloud, no telemetry phone-home, no middle layer that could see your prompts. Your conversation data lives exclusively on your Mac. It's for privacy-conscious Mac users — developers, researchers, anyone handling sensitive information — who want a chat client that's as fast as Notes.app and as private as a local text file.

Quick Setup

Step 1: Get your AVELIN API key

Sign up at avelin.ai and generate an API key. It will look like sk-ave...xxxx.

Step 2: Install Warden

Download from Warden and install for your platform.

Step 3: Configure AVELIN

  1. Open Warden
  2. Go to Settings or API Configuration
  3. Select OpenAI Compatible or Custom Provider
  4. Enter your credentials:
FieldValue
API Base URLhttps://api.avelin.ai/v1
API Keysk-a...
API TypeOpenAI Compatible

Step 4: Add Models

Add your preferred AVELIN models:

avelin-pro
avelin-ultra
avelin-fast
avelin-coding-pro
avelin-coding-ultra
Use caseModelWhy
Daily conversationavelin-proBalanced quality and speed
Complex analysisavelin-ultraFlagship reasoning
Quick questionsavelin-fastLowest latency
Code tasksavelin-coding-pro1M context, deep reasoning

Why AVELIN for Warden?

  • Governed API: Enterprise-grade backend with automatic failover
  • Prompt caching: ~80% cost reduction on repeated prompts
  • Large context: 256K-1M token windows for long conversations
  • Model diversity: 9 specialized models through one API key

Troubleshooting

IssueSolution
"Invalid API key"Verify key starts with sk-ave
Connection failedCheck base URL is https://api.avelin.ai/v1
Slow responsesTry avelin-fast for lower latency
Model not showingAdd it manually in settings