a+Terminal vs CoreAI Zoo

a+Terminal is a privacy-first SSH terminal for working on your own machines from your iPhone — built around tmux and CLI AI coding agents. It's agent-agnostic: run any CLI coding agent, with plug-and-play profiles you can extend yourself. SCROLLING THAT ACTUALLY WORKS Swipes become real mouse-wheel events, so tmux and your agent's output scroll natively, with momentum — no more fighting copy-mode. SPEAK YOUR PROMPTS Push-to-talk dictation inserts straight into the terminal. Recognition runs entirely on your iPhone; audio never leaves the device. NEVER LOSE YOUR SESSION Live Activities and the Dynamic Island show your active sessions at a glance. Tap to jump back in — a+Terminal reconnects and re-attaches your session multiplexer (tmux, zellij, and more) automatically. PREVIEW YOUR DEV SERVER Start a dev server on the machine you're connected to and a+Terminal spots the port — from the URL your framework prints, or from the machine's own listener table. Tap it and the page opens over the SSH connection you already have. It renders in a viewer built for exactly one thing: the single port you forwarded. Every other address is refused, nothing it loads is kept, and it is not a web browser. Reload with a tap, and switch between the mobile and desktop layout to check both. MONITOR YOUR MAC Add a Monitor (VNC) session to watch your own computer's screen sharing right from your phone. Sign in with your macOS account and you land straight on the desktop; pinch to zoom and pan. Pop the screen out into a floating Picture-in-Picture window to keep an eye on it while you use other apps. Turn on Control to tap, drag, and type — enough to unlock a locked Mac and take over. It stays view-only until you do, and it connects only to the machine you enter. (Beta.) ZERO DATA COLLECTION No analytics, no accounts, no crash SDKs, no servers of ours. The App Privacy label says "Data Not Collected" because there is nothing to collect it with. The only network traffic is your own connections to machines you configure. SUPPORT THE APP An optional tip jar lets you support a+Terminal's development. Tips are one-time and unlock nothing — every feature is always available. • ed25519 keys generated on-device, stored in the Keychain (never exportable, never in backups) • Password authentication option, stored in the Keychain only • Trust-on-first-use host key pinning with hard mismatch failures • Sticky Ctrl and C-b (tmux prefix) keys, Esc/Tab/arrows accessory bar • Attach an image or file from your phone over the existing SSH connection • Plug-and-play agent & multiplexer profiles — any CLI agent, tmux, zellij, screen — or define your own, no update required • Multiple simultaneous sessions with server groups • Preview a dev server running on the far end, over the SSH connection you already have — port detection and a single-port viewer • Monitor (VNC): view-only screen sharing for your own Mac, with a floating Picture-in-Picture window and optional touch & keyboard control (beta) • Light, dark, and follow-system themes Privacy Policy: https://aaroncx.github.io/a-plus-terminal/privacy

CoreAI Zoo puts the open-weight AI model zoo on your iPhone, iPad and Mac. Pick a model, download it once, and run it entirely on your device. No account, no server, and nothing you type, say or photograph ever leaves the device. CHAT Stream answers from open language models such as Qwen3, Gemma, LFM2.5, Granite and Mistral, from 0.6B to 30B+ parameters depending on your device's memory. Vision-language models (Qwen3-VL, MiniCPM-V, LFM2.5-VL) answer questions about your photos. VISION Ask questions about an image, read text with on-device OCR, and describe what the camera sees. AUDIO Transcribe speech with Whisper large-v3-turbo, Parakeet and Qwen3-ASR. Talk to an audio-understanding model, generate speech with Kokoro and VoxCPM, and try on-device music generation. CAMERA Live depth estimation, object detection (YOLOX, RF-DETR), action recognition and super-resolution on the camera feed. TOOLS Experimental models such as diffusion-based language models. BENCH Measure load time, tokens per second and memory for any chat model on your own hardware. HOW IT WORKS Every model runs through Core ML on Apple silicon. Weights are downloaded from the Hugging Face Hub on first use over HTTPS, then cached, so the app works offline afterwards. One model is resident at a time to stay within memory limits. New models arrive through a live catalog without an app update. GETTING STARTED Start with Qwen3-0.6B (454 MB) for the fastest first run. Larger models state the free space and memory they need; a recent iPhone Pro or an Apple silicon Mac runs the mid-size ones comfortably. Model weights follow their own licenses, listed on each model's Hugging Face page. The app itself is BSD-3-Clause and part of the open CoreAI model zoo. Report issues and request models at github.com/john-rocky/coreai-model-zoo.