OpenClaw Brings Native App Onboarding to macOS and Easy Local AI to Windows RTX PCs

OpenClaw has released a major onboarding overhaul, shipping a native macOS installer and significantly easier local model setup for Windows NVIDIA RTX PCs.

macOS: install like any other app

Users on macOS can now download OpenClaw and click through a familiar app-installation UI, getting to a working agent without touching the terminal. OpenClaw tests the installer in clean environments to surface the spots where connections and authentication tend to fail during first-run.

Automatic model detection on macOS

Model setup is a common friction point. If you already have working AI access — Claude, Codex, or a local Ollama instance — OpenClaw on macOS can now detect and verify the connection point automatically. Power users can still override and configure manually.

Frictionless local AI on Windows RTX

On Windows PCs with NVIDIA RTX GPUs (GeForce RTX or RTX PRO, 24GB+ memory), OpenClaw goes a step further: it detects the hardware during onboarding and suggests optimized local models to run, without requiring any prior configuration. Once set up, the Windows app manages the local model and llama-server.

This opens local AI capabilities to a much broader group of users, while giving privacy-conscious users private, unmetered inference on their own hardware.

Support for NVIDIA RTX Spark and DGX Station for Windows is planned.

More details are in the official announcement on the OpenClaw blog.