OmniClassify vs pocket/grammer

Transform your iPhone into a powerful machine learning lab. OmniClassify is the complete iOS toolkit for training, deploying, and running AI models entirely on-device. Your data never leaves your device—complete privacy, zero cloud uploads. COMPLETE ML WORKFLOW Import datasets from local folders, videos, Parquet files, or download directly from HuggingFace. Train custom models using CreateML, ONNX, or Vision Transformer architectures. Run real-time inference on images, videos, and live camera feed. Process everything locally with zero data transmission. KEY FEATURES Dataset Management Import from local folders, videos, Parquet files, or HuggingFace (Premium). DICOM medical imaging support. Automatic class detection, train/validation splits, and comprehensive statistics. Model Training (Premium) Train classification and segmentation models with CreateML, ONNX, or ViT. Custom training iterations (unlimited), advanced preprocessing, and progress monitoring. Model Import CoreML and ONNX model support. HuggingFace model downloads with automatic server-side conversion. Thousands of pre-trained models available. Image Classification Batch processing with unlimited capacity (Premium). Multi-model inference, confidence scores, detailed metrics, and result export (Premium). Object Detection YOLO model support with real-time bounding boxes. Multi-object detection, confidence thresholds, and live camera capabilities. Image Segmentation Semantic segmentation with pixel-level masks. Visual overlays and export capabilities (Premium). Video Processing Frame-by-frame AI analysis with custom intervals. Export annotated videos with bounding boxes, labels, and segmentation overlays. Live Camera Inference Real-time classification, detection, and segmentation using iPhone camera. Instant on-device results, works completely offline. DICOM Medical Imaging Full DICOM file support with multi-frame extraction, header reading, and integration with classification/detection models. HuggingFace Integration (Premium) Browse and download thousands of datasets and models. Automatic ONNX/CoreML conversion. Search by task type and license. Advanced Analytics (Premium) Confusion matrices, training metrics, validation curves. Export to JSON, CSV, PDF formats. Automation & Triggers Detection-based automations, Shortcuts app integration, custom notifications, URL opening, and haptic feedback. 100% ON-DEVICE PROCESSING Everything happens locally on your iPhone. Training, inference, analysis, and storage—all without internet or cloud servers. Your images, models, and data remain completely private. No uploads, no tracking, no data collection. PRIVACY FIRST All ML processing on-device. No cloud uploads. Complete offline functionality. No analytics or tracking. HIPAA-compliant workflows for medical imaging. PERFECT FOR ML researchers, data scientists, medical imaging professionals, computer vision developers, students, and hobbyists experimenting with AI. PREMIUM FEATURES Unlock unlimited datasets, unlimited models, unlimited inference, advanced training, exports, HuggingFace integration, and comprehensive analytics. Free tier includes 3 datasets, 2 models, and unlimited inference—perfect for getting started. Start building and deploying AI models on your iPhone today. Complete privacy. Complete control. Complete ML toolkit. Terms of Use: https://www.apple.com/legal/internet-services/itunes/dev/stdeula/ Privacy Policy: https://omniclassify.com/privacy

Agent (AI coding) - Plan mode: conversational clarify phase (capped rounds), then plan generation - Plan lint gate — plans are validated before code mode; bad plans bounce back - Deliberate Plan→Code handoff with banner in Agent tab - Step-by-step plan execution (run remaining / per-step), per-step review cards - Per-card Undo on agent file writes, backed by a shared step-history ledger (max 50 entries, per project) - File tools: create, insert, delete, move, copy — with gated tool selection per message - Image picker + clipboard attachments to chats; streaming responses with haptics - Role models: separate plan/code models, tiered by device RAM, with user overrides Models - Local GGUF inference via llama.cpp (Metal GPU layer probing/splitting, auto thread count, sampling params, grammar-constrained tool calls) - Model browser: recommended / downloaded / HuggingFace catalog with infinite scroll - Download manager, custom model entry, per-model JSON sidecars - BYOK: OpenAI-compatible providers, keys in SecureStore, effort levels, redaction - IDE-compat server: OpenAI/Ollama-style endpoints (/v1/chat/completions, /api/chat, …) so external IDEs can use the phone's models Workspace (mobile or desktop editor) - File tree with open-file tabs (cap, dirty-close prompts), new/rename/copy/delete - Drag-to-move files/folders (long-press to drag) - Syntax highlighting for multiple languages, inline autocomplete suggestions - Project-wide find, diagnostics bar, per-step undo/revert in editor - JavaScript sandbox runner, HTML/app/image/doc preview via WebView - Share/export files, error review screen for pasting errors to the model Projects & connectivity - Project create/import (document picker, clipboard), GitHub repo import (zip codeload with binary/junk filtering), project export/zip - On-device HTTP server + WebSocket with per-install auth token - Desktop client (Vite+React+Monaco) bundled into the app — phone serves the full desktop UI locally; desktop has Monaco editor, terminal panel, search panel, workspace history - api-server backend: projects/files/run-stop/search/import endpoints, LLM agent chat (streaming), model management, preview route, billing/legal