PocketLLM Server

Turn your iPhone into a private AI server.

PocketLLM Server runs language models directly on your iPhone and makes them available to your computers, apps and AI agents through an OpenAI-compatible API on your local network. The iPhone runs the model; your computer runs your code, tools and agents.

Laptop / Agent / App
          |
     OpenAI API  (Wi-Fi)
          v
       iPhone  ->  Local LLM

Quick start

  1. In PocketLLM, open Models and download a recommended model (or tap + to import your own GGUF file from Files).
  2. Tap Load.
  3. On the Server tab, tap Start Server.
  4. Copy the Endpoint (http://<iphone-ip>:8080/v1) and the API key.
  5. From a computer on the same Wi-Fi, point any OpenAI-compatible client at that endpoint.
from openai import OpenAI

client = OpenAI(base_url="http://IPHONE_IP:8080/v1", api_key="YOUR_API_KEY")
reply = client.chat.completions.create(
    model="local-model",
    messages=[{"role": "user", "content": "Say hello in one short sentence."}],
)
print(reply.choices[0].message.content)

The app’s Connection Examples screen gives you this snippet (and PowerShell, curl and JavaScript versions) already filled in with your endpoint and key.

More

PocketLLM runs only while the app is open in the foreground: iOS pauses the server when the app is backgrounded or the phone locks.

This site is open source. Improve this page.