Ollama (on your computer)

Overview

Ollama runs open models on your computer. Loom on your phone then talks to it over your home Wi-Fi. Nothing goes to any company, and no key is needed.

By default Ollama only listens to the computer it is on, so you have to let your network reach it.

Set it up

1. 1
   Install Ollama from ollama.com/download and pull a model, for example ollama pull llama3.2 (pick one that fits your computer’s memory).
2. 2
   Let your network reach it. Set OLLAMA_HOST=0.0.0.0:11434. On a Mac app: launchctl setenv OLLAMA_HOST "0.0.0.0:11434" then quit and reopen Ollama. On Linux: add Environment="OLLAMA_HOST=0.0.0.0:11434" to the service with systemctl edit ollama, then restart it. On Windows: add it as a user environment variable, then quit Ollama from the tray and start it again.
3. 3
   Allow it through the firewall for your private network, on port 11434 (on Windows, add an inbound rule; on Linux, for example sudo ufw allow from 192.168.1.0/24 to any port 11434).
4. 4
   Find your computer’s address on the Wi-Fi, such as 192.168.1.20. Your phone must be on the same network.
5. 5
   Recommended: raise the context window so long chats are not silently cut. Set OLLAMA_CONTEXT_LENGTH=64000 the same way, if your computer has the memory.

Cost. Free. You use your own computer’s power.

Add it to Loom

1. 1
   Open Loom, then Providers in the menu, and choose Add provider.
2. 2
   Pick Ollama (the search box finds it).
3. 3
   Fill in the fields in the table below.
4. 4
   Choose Connect. Loom checks the key and says how many models it found. If it cannot check, it says so, and you can Save anyway and try a message.
5. 5
   Open the model chip below the message box and pick a model.

• Endpoint: http://192.168.1.20:11434 (use your computer’s address). On an Android emulator use http://10.0.2.2:11434.
• Leave the key empty. Loom checks the connection by listing the models you have pulled.

At a glance

Good to know

If something goes wrong

Official documentation

• Ollama documentation