ArcanAI vs PalmiAgent

ArcanAI is a private, on-device AI chat app that runs open-source language models entirely on your iPhone or iPad. No accounts. No subscriptions. No cloud. No internet required. Import any GGUF model from Hugging Face and start chatting — every conversation happens locally, powered by hardware-accelerated inference using Apple's Metal framework. Your messages never leave your device. KEY FEATURES - Run AI models 100% offline — on a plane, off the grid, anywhere - Import any GGUF model from Hugging Face — choose the size and capability that fits your device - Store and switch between up to 5 models at once - Analyze documents — upload PDFs, spreadsheets, text files, CSVs, and more - GPU-accelerated inference for fast, streaming responses - Full conversation history with multiple independent chats - Edit messages and regenerate responses - Customize temperature, max tokens, and system prompts - Beautiful dark UI with markdown-rendered responses - Text-to-speech support SUPPORTED MODEL FAMILIES Llama, Gemma, Mistral, Phi, Qwen, and any ChatML-compatible model. RECOMMENDED STARTER MODELS - Llama 3.2 1B (~570 MB) — great for older devices - Gemma 2 2B (~1.0 GB) — balanced performance - Qwen 2.5 3B (~1.5 GB) — multilingual and code capabilities PRIVACY FIRST Zero data collection. Zero telemetry. Zero crash reporting. Zero cloud sync. Your conversations and model files stay on your device — always.

PalmiAgent is an AI agent workspace for iPhone. It brings conversations, project files, model configuration, multimodal input, web research, and tool execution into one task environment—so you can do more than chat and keep real work organized. PROJECT-BASED CONVERSATIONS Organize work in projects and threads. Chat history, attachments, generated files, OCR output, Python logs, and web captures stay connected to the active workspace and can be revisited or exported. FLEXIBLE MODEL CONNECTIONS Create model plans with primary, multimodal, and lightweight roles. Connect GLM/Z.AI, DeepSeek, local services, or custom OpenAI-compatible endpoints. A thread can temporarily override the global plan. IMAGES AND ON-DEVICE OCR Send images for model-assisted understanding. For text extraction, bundled PP-OCRv6 Tiny assets can run OCR on device and save both readable text and structured recognition results. FILES AND PYTHON Read, write, organize, and preview workspace files. An embedded CPython runtime and curated pure-Python packages support calculations, data handling, parsing, and workspace automation. WEB RESEARCH Search the web, read static pages, fetch sources in batches, and preview pages inside the app. Standard chat, Goal Mode, and Deep Research Mode support different styles of work. SKILLS AND CONTROLLED TOOLS Import and enable skill packages to give the agent task-specific instructions. Tools that write files, access personal data, or perform system actions follow your authorization settings. With your permission, PalmiAgent can work with iOS capabilities such as Calendar, Reminders, Contacts, Location, Photos, Camera, Notifications, Speech, and Maps. PRIVACY Sessions and workspace files are stored on your device by default, and API keys are kept in the system Keychain. PalmiAgent does not operate its own cloud model server. When you use a model or search service, relevant content is sent to the third-party provider you configured and is governed by that provider’s terms and privacy policy. PLEASE NOTE PalmiAgent does not include general-purpose model weights or a hosted model service. You need to provide a compatible model service, account, or API key. Some features require network access, third-party service availability, and the corresponding iOS permissions.