SwiftServer vs Pocket AI Lab

SwiftServer is an SSH client and server monitor for iPhone, iPad, and Mac. It is for developers, sysadmins, and homelab owners who run Linux, macOS, or Windows servers and want a terminal, live status, Docker, and SFTP in one app. Nothing is installed on the server. SwiftServer connects over the SSH access you already have. No agent, no daemon, no account. MONITORING - CPU per core, load averages, and temperature - Memory as used, cached, free, and swap - Live process table, sortable and searchable - Network speed and totals, with vnStat history by hour, day, month, and year - Storage per volume with disk I/O speed, latency, and IOPS - NVIDIA GPU: utilization, memory, fan, temperature, power, and compute processes - Public IP and location per server. Global View maps every server with live latency. - Linux, macOS, and Windows on one dashboard - Widgets, Live Activities, and the Dynamic Island TERMINAL - SSH and Mosh sessions. Mosh keeps the shell alive through network changes and sleep. - Keyboard bar with Esc, Tab, Ctrl, Alt, arrows, and ^C on iPhone and iPad. Hardware keyboards work too. - Snippets: save a command once, run it in any session, or across many machines with a result per machine - 518 color themes with separate light and dark picks, nine bundled monospace fonts, and custom font import - Mac: Local Shell opens a terminal on your own Mac in the same window DOCKER - Speaks the Docker Engine API over the SSH connection. No Docker port is exposed. - Overview: engine status, counts, reclaimable space, and live CPU and memory across containers - Containers: logs, processes, inspect. Create, start, stop, restart, pause, kill, remove. - Compose projects: up, down, start, stop, restart, pull, and logs - Images: pull with progress, remove, prune. Volumes and networks: inspect, remove, prune. Live event stream. SFTP - Browse, upload, download, rename, delete, and create files and folders - Copy files from one server to another without routing them through your device - Quick Look preview and a text editor with save NETWORKING - Port forwarding: local, remote, and dynamic SOCKS rules with auto start and a built-in browser on iPhone and iPad - Jump hosts: reach machines behind a bastion through one hop or a chain, each with its own credentials and host key check - Connection proxies: SOCKS5 and HTTP CONNECT profiles, per machine or per forwarding rule KEYS AND SECURITY - Password, SSH key, and keyboard-interactive login, including two-factor prompts - Generate Ed25519, ECDSA, or RSA keys on the device, or import existing keys - Host key fingerprint on first contact. A changed host key stops the connection. - Secrets stay in the device Keychain, behind Face ID, Touch ID, or passcode if you choose - Built on Traversio, our open-source SSH engine written in Swift ICLOUD SYNC AND PRIVACY - Machines, identities, keys, snippets, forwarding rules, proxy profiles, and trusted host keys sync through your own iCloud. Passwords and private keys go only through iCloud Keychain, end-to-end encrypted. - No accounts and no SwiftServer servers. What you save stays on your devices and in your iCloud. - The only outbound request is the IP lookup, sent from your server to ipinfo.io. The result is stored on your device only. FREE AND PREMIUM The core of SwiftServer is free: saved machines and keys, live server status, the full terminal with Mosh and snippets, SFTP, port forwarding, Docker browsing, widgets, and Global View. Premium adds unlimited monitored servers, iCloud sync, unlimited forwarding sessions, jump hosts, proxies, Docker management, terminal themes and fonts, SFTP preview and editing, and city, organization, and map pin for IP location. Premium is a monthly or yearly subscription, each with a free trial, or a one-time lifetime purchase. One purchase covers iPhone, iPad, and Mac. Support: support@gitswift.com Terms of Use: https://swiftserver.app/terms Privacy Policy: https://swiftserver.app/privacy

Pocket AI Lab scans your device, tells you honestly which models will actually fit, downloads them straight from Hugging Face, and lets you chat with them fully offline. Multi-backend inference — MLX, llama.cpp (GGUF), Core ML, and Apple Intelligence, switchable at runtime Multimodal — text, images and video frames (vision models), speech-to-text (Whisper) Import any model by link — paste a Hugging Face URL, pick a quantization, get a fit verdict before a single byte downloads Honest memory budgets — recommendations are computed from the real per-process memory limits of your device, not marketing RAM numbers Live metrics — CPU, RAM and thermal pressure while the model is thinking Streaming chat — token-by-token, with full conversation history on every backend