mycellm vs MChatP

Mycellm turns your iPhone or iPad into a node on a distributed AI inference network. Run open LLMs on your own hardware — no account, no telemetry, no cloud middleman. Use it three ways, and own every layer. LOCAL — ON YOUR DEVICE Load a model and run inference entirely on-device. Fully offline, airplane-mode friendly. On-device Privacy Guard scans prompts for secrets before anything hits the wire. TRUSTED DEVICES — YOUR OWN NETWORK Point your phone at a node you trust — your Mac Studio, your iPad Pro, your homelab — and run models far bigger than any single device could hold. Your phone stays light; a machine you own does the heavy lifting. No cloud. PUBLIC — THE FREE NETWORK Tap the public mycellm net: open models served by GPUs worldwide. Earn credits by serving spare compute, spend them to run larger models. Peer-to-peer, no subscription. FEDERATE Stand up your own network with its own trust policy and credit ledger, mint invite tokens, and bridge trusted networks with cross-network routing. You decide membership, models, and data flow. UNLEASH THE iPAD PRO An iPad Pro (M4/M5) is a serious AI machine that sits idle most of the day. Put it to work: more unified memory runs larger models, Metal delivers 30+ tok/s on multi-billion-parameter models. Serve your iPhone, your team, or the network — first-class on any device that runs a model. FOR PEOPLE WHO RUN THEIR OWN STACK - Serves the same chat-completions REST API your tools already speak — point curl, Claude Code, or your own clients at your device. - Vision models: attach an image and ask, on-device. - Live dashboard: tok/s, credits, peers, node health. - Auto hardware-fit: picks quantizations that match your RAM. - Open source, Apache 2.0. Audit it, fork it, run it. No black boxes, no lock-in. It's BitTorrent for LLMs — every node both uses the network and makes it stronger. Full docs and use cases at mycellm.dev. Bring your own models, keep your data, and run AI on your terms.

MChatP is a chat client for the AI server you run yourself. You choose the computer, the server software and the models: LM Studio, or any OpenAI-compatible server such as Ollama, oMLX or llama.cpp, with the open models you installed. MChatP hosts no model and runs no backend: it talks to your server, at home over your local network or from anywhere over a private network such as Tailscale, and shows the replies. Using MChatP requires a computer already running one of these servers (third-party software); the in-app Help walks through the setup step by step. YOUR SERVER, YOUR MODELS • Any model your server offers, from small and fast to large ones no phone could run, switched in one tap • LM Studio through its native API, with the MCP tools of your mcp.json: your files and tools stay on your computer, the server runs them • Ollama, oMLX, llama.cpp and any server exposing /v1/chat/completions, token optional • Several server profiles (MChatP Pro) MANAGE MODELS REMOTELY (MCHATP PRO) Load or unload a model on your LM Studio computer, or download a new one there from its catalog ID or a Hugging Face link, with live progress. PHOTOS AND DOCUMENTS • Show up to four photos to a vision model • Attach a PDF, text, Markdown, JSON or CSV file: its text is extracted on your device and sent with your message, so it works with any model WATCH IT THINK • The model's reasoning, folded above the answer • MCP tool calls as they start, succeed or fail • Model loading and prompt progress, then speed in tokens per second • Developer Mode adds time to first token, token counts and the full detail of each tool call (MChatP Pro) SYSTEM PROMPTS AND CHAT TOOLS A system prompt to steer the model (one included, a library with MChatP Pro); stop or regenerate a reply, copy, rename, search your history, keyboard shortcuts (⌘N, ⌘↩, ⌘K). HANDS-FREE • Ask MChatP from the action button or by voice; the answer comes back without opening the app • Start a conversation or test your server from the system's shortcuts PRIVATE BY DESIGN • Server profiles, tokens, conversations and attachments stay on your device • No account, no cloud sync, no analytics • Tokens in the device keychain; optional PIN lock with biometric unlock • Transparent about AI: you always know you are chatting with an AI model, the one you loaded on your own server, as the European AI Act requires MCHATP PRO, ONE-TIME PURCHASE Several server profiles, remote model management on LM Studio, a library of system prompts, Developer Mode and conversation export, unlocked with a single non-consumable purchase, charged at confirmation; the price is shown in the app. No subscription, nothing renews. After reinstalling or changing devices, Restore Purchases unlocks it again at no extra cost. MChatP is an independent app, not affiliated with the makers of LM Studio, Ollama, llama.cpp, oMLX or Tailscale, trademarks of their respective owners. Privacy Policy: https://precellence.icu/apps/mchatp/privacy_policy.html Terms of Use: https://precellence.icu/apps/mchatp/terms_of_use.html Support: support@precellence.icu