Tutorial 2: Your First AI Conversation in 5 Minutes

Think of setting up Hermes like turning on a new smartphone: you don’t need to read the manual cover-to-cover—you just pick it up, choose your network, and make your first call. This tutorial gets you from zero to a working AI conversation in about five minutes, so you can see Hermes actually respond before you worry about any fancy features.

Step 1: Install Hermes (the 30-second part)

If you’re on macOS or Windows, the easiest route is the Hermes Desktop installer—just download it from hermes-agent.nousresearch.com and run it. That gives you both the desktop app and the command-line tool.

Prefer the command line only? Open your terminal (or PowerShell on Windows) and run one of these:

Linux / macOS / WSL2 / Android (Termux):

curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash

Windows (native, in PowerShell):

iex (irm https://hermes-agent.nousresearch.com/install.ps1)

After it finishes, reload your shell so the hermes command is available:

source ~/.bashrc   # or source ~/.zshrc

That’s it. No complicated dependencies, no manual path editing.

Step 2: Choose Your Provider (the “network” for your AI)

Hermes doesn’t come with a built-in brain—it connects to one. The single most important setup step is telling Hermes which AI provider to use. Run this command:

hermes model

This launches an interactive menu that walks you through your options. If you’re brand new, the easiest path is Nous Portal—one subscription gives you access to 300+ models plus extra tools like web search and image generation, all without juggling multiple API keys.

Don’t overthink this. Pick a provider, follow the prompts, and save the config. You can always switch later.

Step 3: Have Your First Real Conversation

Now for the fun part. With your provider set, just start chatting:

hermes

Type something simple: “Hello, what can you do?” or “Write a haiku about coffee.” If Hermes responds with something sensible, congratulations—you’re officially up and running.

Rule of thumb: If Hermes can’t complete a normal chat, don’t add more features yet. Get one clean conversation working first. That’s your foundation.

Step 4: Know What to Do If Something Breaks (the 2-minute fix)

Most setup problems come down to three things:

Provider not configured – Re-run hermes model and double-check your API key or endpoint.
Wrong model name – If you chose a custom or local endpoint, verify the exact model name and context length.
Shell not reloaded – If hermes isn’t recognized, run source ~/.bashrc again.

If you’re using a local or self-hosted model, test the endpoint separately before blaming Hermes. And if you’re on Android/Termux, check the dedicated Termux guide for platform-specific notes.

Step 5: What NOT to Do Yet

It’s tempting to immediately add a Telegram bot, cron jobs, or voice control. Resist. The official docs are blunt about this: if Hermes cannot complete a normal chat, do not add more features. Layer on gateway, skills, or routing only after you’ve had a few clean conversations.

Summary

You just installed Hermes, picked a provider, and had your first AI conversation—all in under five minutes. The commands to remember:

hermes setup – guided initial setup
hermes model – choose or change provider
hermes – start chatting

Next up: In Tutorial 3, we’ll show you how to turn Hermes into an always-on assistant by connecting it to Telegram or Discord—so you can chat with it from your phone, even when your laptop is closed.