Launching today

Localmee

Launching today

Private AI chat: files, voice, vision. 18 local models.

7 followers

Private AI chat: files, voice, vision. 18 local models.

7 followers

Visit website

AI Chatbots

Localmee is a private, offline AI chat workspace for iPhone, iPad, and Mac. Run 18 curated open-source models and counting, including Qwen3.5 and Gemma 4, powered by Apple’s on-device MLX framework. Chat with files, images, and voice. Summarize PDFs, EPUBs, DOCX, and 500+ page books. Use OCR, vision, and ready-to-run prompts for work, study, writing, and coding. Localmee recommends models for your device. On Mac, connect your own Ollama models. No backend, no tracking, no account required.

Free Options

Launch tags:Productivity•Privacy•Artificial Intelligence

Launch Team / Built With

SureThing.io — Your first AI hire - for owners who don't "do tech"

Your first AI hire - for owners who don't "do tech"

Promoted

Maker

📌

Hey Product Hunt 👋 I’m Tuba, and I’ve been building Localmee together with my friend David. We believe personal AI is moving toward a new layer of computing: not just something you open in a browser, but something that lives with you across your devices, understands your files and context, works without constant internet access, and stays private by default. The future of AI should not be only centralized, rented, and cloud-dependent. Some of it should be local, personal, and fully under your control. That’s the idea behind Localmee. Localmee is a private, offline AI chat workspace for iPhone, iPad, and Mac. It runs curated open-source models directly on your device, including Qwen3.5 and Gemma 4, powered by Apple’s on-device MLX framework. On Mac, we also support Ollama, so power users can connect their own local models and use Localmee as a private AI interface for the models they already run. We think this is an important kind of freedom: your device, your models, your workflow. What you can do with Localmee today: • Chat with local AI models • Ask questions about PDFs, EPUBs, DOCX, notes, and long books • Analyze images and extract text with OCR • Use voice chat • Start faster with a built-in prompt library • On Mac, connect your own Ollama models • Work offline, with no account, no backend, and no tracking Localmee currently supports 18 curated open-source models, and we’re adding more as better on-device models become available. But we see this as the beginning, not the finish line. Our roadmap is focused on turning Localmee into a deeper private AI layer for your Apple devices: smarter document workflows, richer model discovery, better voice experiences, more visual understanding, stronger prompt tools, and new ways to make local AI feel useful in everyday work instead of hidden behind technical settings. In short: we want Localmee to become the place where private AI, open models, your files, and your daily workflow meet. We built this for people who want AI to feel personal, useful, and private by default. We’d love your honest feedback: Where would private, offline AI be most useful in your daily workflow? Thanks for checking it out. Tuba

Report

1d ago

Congrats on the launch, wishing you all the best with it!
Just downloaded it and have been playing with it for a bit. The fact that everything runs locally on the device is a huge plus for me, response times are surprisingly snappy and I love that I can use it with no signal. Voice quality is also better than I expected for on-device TTS.

Quick question: any rough timeline for adding new models, or is there a public roadmap I can follow? Curious what's coming next on the model side.

Great work, will keep using it.

Report

10h ago

Maker

@sadegazoz Thank you so much, this really means a lot on launch day!
Glad the on-device experience and voice landed well for you, those were two of the hardest things to get right.

On the roadmap side, me and @davidcs have already started working on the next chapter: pushing Localmee to its full potential as a truly private, fully offline assistant that can handle a lot more than chat. Not ready to share all the details yet, but there is a lot coming.

Thanks again for trying it out and taking the time to write this.

Tuba

Report

10h ago