Inception Labs Docs

Appearance

OpenClaw Integration ​

Fresh

Use Mercury 2 with OpenClaw, a locally-running personal AI assistant for messaging platforms (WhatsApp, Telegram, Discord).

Setup Steps ​

1. Set Environment Variable ​

bash
export INCEPTION_API_KEY="your_api_key_here"
# Add to ~/.zshrc or ~/.bash_profile for persistence
bash
set INCEPTION_API_KEY="your_api_key_here"

2. Configure Provider ​

Edit ~/.openclaw/openclaw.json to add Inception:

json
{
  "providers": {
    "inception": {
      "base_url": "https://api.inceptionlabs.ai/v1",
      "api_type": "openai-completions",
      "models": {
        "mercury-2": {
          "context_window": 128000,
          "max_output": 16384
        }
      }
    }
  },
  "mode": "merge"
}

3. Set Default Model ​

Use the CLI to set Mercury 2 as your primary model.

4. Restart Gateway ​

Restart the OpenClaw gateway to load the new provider.

5. Verify ​

Run list and status commands to confirm the Inception provider appears.

Runtime Features ​

  • Mid-session model switching: Use /model inception/mercury-2 in any chat
  • Fallback model: Add Mercury as a backup via CLI

Diagnostics ​

Run built-in health checks and gateway status commands for troubleshooting.

See Also ​