Local Setup
TightSpot is private by default. Your material stays on your device and Coach analysis is the only thing that ever touches a network (and even that is private). If you want to go further and keep everything on your own hardware with no external connections at all, this guide is for you.
Before you start
Local analysis runs on your computer, not your phone. Your phone and computer need to be on the same wifi network for this to work. It will not work at a venue unless you bring your laptop on stage with you. We don't recommend this.
Performance depends on your hardware. On a modern Mac with Apple Silicon the experience is comparable to the default cloud option. On older hardware, analysis that takes seconds in the default mode may take a minute or more. A machine from 2020 or newer with at least 8GB of RAM is recommended.
The simplest option today is Ollama, a free open-source tool that runs AI models locally on your machine. Choose your platform:
Download the installer from ollama.com and run it. That's it.
Download the installer from ollama.com and run it.
Run this in your terminal:
curl -fsSL https://ollama.com/install.sh | sh
Open a terminal (Mac/Linux) or Command Prompt (Windows) and run:
ollama pull llama3.2:3b
This downloads about 2GB. It only happens once. The 3b model runs well on most hardware. If your machine has a dedicated GPU or Apple Silicon, you can try llama3.2:7b for better results.
Your phone needs to know where to find the AI engine on your network.
Go to System Settings, click Wi-Fi, then click the network you're connected to. Your IP address is listed as "IP address."
Open Command Prompt and type ipconfig. Look for "IPv4 Address" under your active network adapter.
Run ip addr show in a terminal. Look for the inet address on your active network interface (usually eth0 or wlan0).
Your IP address will look something like 192.168.1.42. Write it down — you'll need it in the next step.
Open TightSpot on your phone and go to the Green Room. Tap "Change analysis source" and select Advanced.
Enter your computer's IP address in the URL field using this format:
http://192.168.1.42:11434
Replace 192.168.1.42 with the address you found in Step 3. Leave the model name as llama3.2:3b unless you pulled a different model in Step 2.
Tap Save.
Go to the Coach tab and tap "Read My Material." If it works, you're done.
If it times out or fails, see the troubleshooting section below.
If something goes wrong
Request times out
Make sure your phone and computer are on the same wifi network. A phone on mobile data and a computer on wifi won't see each other.
Connection refused
The AI engine may not be running. Open a terminal and run ollama serve to start it manually.
Works on Mac, fails on Windows
Windows Firewall may be blocking port 11434. Add an inbound rule to allow TCP traffic on that port, or temporarily disable the firewall to test.
IP address changed
Local IP addresses can change when your router reassigns them. Check your IP address again (Step 3) and update the URL in TightSpot's Green Room.
Still stuck
Email [email protected] with your platform (Mac/Windows/Linux) and the error you're seeing.