The AI stack on iPhone & iPad.

On iOS you can't run the command-line AI tools themselves — Apple doesn't allow it. So your phone plays two roles: a chat client (the official apps) and a remote terminal into a real computer (an SSH client).

Why this first: for many people the chat apps are all they need. And the SSH client is the key that unlocks every "power" step below — it's how your phone reaches a machine that can run the tools.

From the App Store, install Claude, ChatGPT, and Google Gemini. Sign in to each with the matching account. This gives you full chat anywhere — on the train, in line, in bed. For a lot of people, this is the whole stack.

For real command-line power, you need an app that can connect into another computer's terminal. Two good choices on the App Store:

• Blink Shell — excellent, polished, paid.
• Termius — has a free tier to get started.

Either one lets you log into a Mac, PC, droplet, or Proxmox box and type commands as if you were sitting at it.

Anthropic's AI coding agent — the same kind of assistant that built this page. It reads and edits real files and runs commands. You can't install it on the iPhone, but you can drive it on another computer through your SSH app.

Why you want it: it does actual multi-step work — build a site, fix a script, organize files — not just chat.

Set up Claude Code on a Mac, PC, droplet, or Proxmox box using those guides. Then, from your SSH app on the phone, connect to that machine and run:

claude

That's it — you're now talking to Claude Code on the remote machine, from your phone, in plain English.

OpenAI's command-line coding agent. Same approach as Claude Code on iOS — it runs on a real computer, and you reach it from your phone over SSH.

Why you want it: variety. Different models are stronger at different things; having both means you're never stuck.

Connect via SSH to a machine that has Codex installed, then run:

codex

Google's AI. On iOS you have an easy path and a power path — pick whichever fits the moment.

Why you want it: free headroom. Great for big, sprawling tasks before you spend on the others.

Use the Google Gemini app from the App Store (you installed it in Step 1). Sign in with your Google account and chat away.

SSH into a machine that has the Gemini CLI installed, then run:

gemini

A coding agent from Nous Research — "the agent that grows with you." Same idea as Claude Code, Codex, and Gemini, with two twists: it's self-improving (it learns your preferences over time), and it's model-agnostic — you point it at whichever brain you want (Nous, OpenAI, Anthropic, OpenRouter, and more). Same approach as Claude Code on iOS: it runs on a real computer, and you reach it from your phone over SSH.

Why you'd add it: it's a fourth tool in the rotation and the most model-flexible of the bunch — handy for staying un-locked-in (see the future-proofing principles).

You cannot install Hermes on an iPhone or iPad — Apple doesn't allow it. Instead, install it on a Mac, PC, or droplet using those guides. Then, from your SSH app (Blink or Termius, over Tailscale), connect to that machine and run:

hermes --tui

A private network (a "mesh VPN") that connects all your devices to each other securely — your phone, your Macs, your PCs, your NAS — as if they were in the same room, from anywhere in the world.

Why you want it: this is what makes Steps 2, 3, and 6 actually usable on the go. With Tailscale, your SSH app can reach any of your computers by name — and nothing is exposed to the public internet.

Install the official Tailscale app from the App Store. Sign in with the same account you used on your other devices — that's what joins them all to one private network.

Ollama runs AI models locally — free, private, offline. On iOS the trick is to run it on a real machine and use your phone as a client.

Why you want it: no per-use cost, nothing leaves your control, and one beefy machine at home can answer from your phone anywhere.