API Keys Setup

RespectASO Pro requires an API key from at least one LLM provider to power AI features. Every provider is model-agnostic: you type any model ID the provider offers, click Test model, and the model is usable once the built-in compatibility test passes - so new models work the day they ship, with no app update.

OpenAI

1. Go to platform.openai.com and sign in (or create an account)
2. Navigate to API Keys in your dashboard
3. Click Create new secret key
4. Copy the key (starts with sk-)
5. In RespectASO, go to Settings → AI
6. Paste the key in the OpenAI field, type any OpenAI model ID (for example gpt-5.5), click Test model, then save

Anthropic

1. Go to console.anthropic.com and sign in
2. Navigate to API Keys
3. Click Create Key
4. Copy the key (starts with sk-ant-)
5. In RespectASO, go to Settings → AI
6. Paste the key in the Anthropic field, type any Claude model ID (for example claude-sonnet-4-6), click Test model, then save

Google Gemini

1. Go to aistudio.google.com
2. Click Get API key
3. Create a key in an existing or new project
4. Copy the key
5. In RespectASO, go to Settings → AI
6. Paste the key in the Google Gemini field, type any Gemini model ID (for example gemini-3.5-flash), click Test model, then save

OpenRouter

1. Go to openrouter.ai/keys and sign in (or create an account)
2. Click Create Key
3. Copy the key (starts with sk-or-)
4. In RespectASO, go to Settings → AI
5. Paste the key in the OpenRouter field and save

OpenRouter model IDs use the format vendor/model-name (for example deepseek/deepseek-chat) - free :free models work too. You can browse compatible models at openrouter.ai/models. After entering a model ID, click Test model - the model can only be used after the compatibility test passes. If the test fails, RespectASO tells you the model is not compatible and shows why. The Validate button checks your key for free - no credits are consumed.

OpenRouter routes your request onward to the host of the model you select, giving you access to hundreds of models - including budget options like DeepSeek - through a single key.

Custom endpoint (any OpenAI-compatible service)

Use any AI service that offers an OpenAI-compatible API - OpenCode Zen, Groq, and similar gateways or proxies - without waiting for RespectASO to add it by name.

1. Get an API key from the service's dashboard
2. Find the service's OpenAI-compatible API address in its docs - it usually ends in /v1 (for example https://opencode.ai/zen/v1 for OpenCode Zen, or https://api.groq.com/openai/v1 for Groq)
3. In RespectASO, go to Settings → AI and choose Custom endpoint
4. Paste the address into Base URL, type the model ID exactly as the service names it, and paste the key into the Custom endpoint field
5. Click Test model, then Save Settings

The test checks the key, the Base URL, and the model in one live call, and the model can only be used after it passes. Changing the Base URL later invalidates the test, so re-run it after any change. Your requests go directly from your Mac to the service whose Base URL you enter - make sure you trust it.

Choosing a Provider

Every option produces excellent results. Here's a quick comparison:

You only need one provider configured. You can switch between them anytime from Settings → AI.