MLX vs Hermes Redde

MLX-Serve is a private AI studio that runs entirely on your iPhone. Every conversation, voice recording, and generated image is processed on-device by open-weight models — nothing you type, say, or create ever leaves your phone. There is no account to create, no subscription, and no tracking of any kind. MUSIC Generate songs locally using your device ! You can also play it back in the background and use MLX-Serve as a player for your generated music on the go. CHAT Talk with capable open language models (Google Gemma 4) right on your device. Ask questions, brainstorm, draft messages, get things explained. Some models can even understand photos you attach. CALL Tap Call inside any conversation to talk to the AI hands-free, like a phone call. Your speech is transcribed on-device and the reply is read back aloud. Speaker and handset modes are both supported, and the transcript is saved to your chat. VOICE CLONE Record about eight seconds of your voice and have the AI speak text in your own voice, using an on-device text-to-speech model. Great for narration and fun. IMAGES Generate images from a text prompt with an on-device diffusion model (FLUX.2). Tap a result to zoom in, save it to your photo library, or share it. FULLY PRIVATE, FULLY OFFLINE - Runs 100% on-device using Apple MLX and the Metal GPU. - No account, no analytics, no ads, no tracking. - The network is used only to download the AI models you choose from Hugging Face. After a model is downloaded, that feature works offline. REQUIREMENTS Models are multi-gigabyte downloads and on-device AI is demanding, so MLX Chat requires a recent, high-performance iPhone. Download over Wi-Fi is recommended. You pick which models to download and can delete them anytime to reclaim space. Open models are provided under their respective licenses (Gemma, Apache 2.0). MLX Chat is an independent app and is not affiliated with Google or Apple.

Redde is a complete iPhone and iPad client for Hermes, the open-source AI agent, with CarPlay and an Apple Watch app. Talk or type, watch it think and use tools, approve what it runs, and give it a crew: bots with faces and voices of their own, alone or in group chats. You bring the backend. Redde has no account and no analytics, and your conversations go only to the server you configured. BOTS • Each Hermes profile is a bot with a face, a voice and a chat of its own. Start one from a template (Researcher, Writer, Coder, Home, Planner) and make its job, personality, model, face and tools your own. • Type @ to hand work to another bot. Put two to six in a group chat and they answer in turn; talk to the room and each answers in its own voice. • Give a bot a schedule and its answer arrives in its chat, with a notification, even with Redde closed. Mute any chat for a while. • Ask a bot from Siri, CarPlay, the Apple Watch or a spoken word. TALK, DON'T TYPE • Tap the mic, ask, hear the answer, or show it what the camera sees. Speech is recognized on your iPhone; the reply is spoken by the built-in voice, a Kokoro server or your Hermes. • Hands-free mode keeps the conversation going until you say "that's all". Talk over an answer, or type, to steer it. • In CarPlay, ask through the car's mic and speakers, in a new chat, a recent one or a bot's. On Apple Watch, raise your wrist and ask. • "Hey Siri, ask Redde…" works from the Lock Screen and AirPods. A Control Center button, a widget or the Action Button starts listening. • Spoken prefixes: a word you chose adds a text prefix, switches model or goes to a bot: "Opus, review this diff". • Siri takes notes for your agent; Shortcuts can ask it things. SEE THE AGENT WORK • Watch reasoning and tool calls live, with a Live Activity; lock the phone and the reply carries on. • The list shows what's working, what needs you and what's unread. • Approve a risky command or answer the agent's question from a notification, out loud, or with a nod in AirPods. Pair your Hermes and notifications arrive with Redde closed, encrypted end to end. A FULL CLIENT, NOT JUST A MIC • Conversations live in your gateway's session ledger: start on the phone, continue on the desktop, Telegram or the CLI. Search inside them, pin, rename, fork, archive, or go incognito. • Projects, scheduled jobs and the Kanban board. • Phone tools: your agent can use your iPhone's calendar, reminders, contacts, photos and files, set alarms, write on a widget, and start calls and texts, each only if you allow it. • Rich replies: Markdown, code, tables, Mermaid diagrams, math, and cards with maps, places and the weather. Pictures, PDFs and audio open in place. • Select part of a reply and ask about it. Regenerate, edit and resend, or switch models with /model. • Attach photos, videos, files or a day's calendar, or share from any app. • Skills, tools, memory and context files; on the server, MCP servers, logs, updates, files and an SSH terminal. YOUR BACKEND, YOUR RULES • Connect to Hermes through the Hermes Dashboard or the Hermes API, on your LAN, over a tailnet, or behind Cloudflare Access, Pangolin or a reverse proxy. Save several servers, see all their chats in one list, and sync them with iCloud. Sign in with a password or a browser. • Or go straight to any OpenAI-compatible endpoint: llama.cpp, llama-swap, vLLM or Ollama on your hardware, or a hosted provider with your key. • Set up by QR code or link. Credentials stay in the Keychain. • Face ID lock, seven themes with your own colors, app icons and voice orbs, iPad split view, iPhone Duo, keyboard shortcuts. • The one service we run is an optional relay for notifications, which can't read them. REQUIRES A Hermes agent (github.com/NousResearch/hermes-agent) or an OpenAI-compatible endpoint; bots need Hermes. Redde includes no model and no AI service of its own. Hermes is an open-source project by Nous Research. Redde is an independent client, not affiliated with it.