Guides and stories on on-device AI, LM Studio, Ollama, Personas, and private chat — written by the maker of LM Mini.
Widgets, Shortcuts and Siri let you use LM Mini and your local AI without opening the app first: recent chats and Personas on the Home Screen, questions out loud, and ready-made Shortcuts.
Memory in LM Mini lets a local AI remember things about you from one chat to the next. The AI decides what to keep, the memories stay on your phone, and you can delete any of them.
Image generation in LM Mini lets you ask for a picture from the chat you are in. AUTOMATIC1111 or ComfyUI on your computer makes it, or your iPhone does.

Web search in LM Mini lets a local AI chat look up current info while you stay in the thread.

Compare up to four local models on the same prompt with LM Mini Arena (free) — or run a multi-model Group Chat room (Pro) for debate and specialized roles. Works with LM Studio, Ollama, Jan, oMLX, Unsloth, and on-device.

Using Ollama from your phone means Ollama keeps running on your computer, and your phone is where you chat.

LM Studio does not need internet to chat once a model is on the computer. Neither does Ollama. You need internet to download a model.

On-device models in LM Mini run on your phone, not in the cloud. Smaller quantized files, usually Q4, tend to feel best (just my honest opinion).

Voice chat in LM Mini lets you talk to local AI out loud. Speech-to-text turns what you say into the prompt, text-to-speech reads the reply, including…

Private AI means my chats stay on hardware I control, not a random company's servers.

LM Mini Connect for Mac is deprecated. It keeps working through the end of 2026. The replacement is LM Mini Home — keep working on phone and Mac with realtime sync. Windows and Linux Connect stay current.

Group Chat in LM Mini is one thread with more than one AI in it. Each can be a Persona you already saved, or a different model.

Personas in LM Mini are saved AI characters with a name, look, and instructions, plus an optional model, voice, and memories.

LM Studio remote access from a phone is using the model on your desk from the couch and on the same Wi-Fi, or from outside the house with LM Mini Home on Mac…

On-device AI means the model runs on your phone, not in the cloud. You download it once, then you can chat offline, and your messages stay on your hardware.

LM Mini is a local AI and private AI chat app for iPhone and Android. I built it so I was not stuck at the PC, or carrying it around, just to send a few…