What’s New in Version 0.4.0: • All-New Plan Mode: Turn ambitious goals into structured, actionable execution roadmaps. Supports Multi-Domain Planning across Software Projects, Executive Schedules, Personal Finance, Habit Tracking, and Learning Curricula—with auto-generated milestone tables and progress charts. • The Brain Tab & Cognitive Memory: Experience our living on-device cortical visualizer with organic tethered harmonics. Features a 3-tier persistent memory engine (Working Context, Active Micro-Experts, and Episodic Knowledge) with automated research web ingestion. • Inference Optimizations: Thread affinity scheduled exclusively to Apple Silicon Performance cores for maximum tokens/second. Includes prefix KV caching and real-time streaming token interception. • Compatibility & Engine Upgrades: Native tool-calling protocol for Liquid Foundation Models (LFM), dynamic LoRA adapter attachment, and the new iRun-Link workstation bridge with 7-digit LAN pairing verification. • Cleaner Startup Experience: Chat history sidebar and model selector now default to collapsed on launch for a calm, distraction-free interface. Your open/closed preferences automatically persist across sessions. • Stability Fixes: Completely eliminated tab-switching index errors and gesture translation runaway. Custom LLM on the way, built fully from the ground up.

The "Meltdown" Update & Remote Compute Power! Version 0.3 brings major platform upgrades, expanded local memory capabilities, and the ability to link your Apple device to external hardware for heavy-duty inference. iRun Link (Remote Compute Client) Custom Server Connection: Connect iRun directly to your custom Windows or Linux server to offload heavy inference tasks to remote CUDA hardware over your local network. Meet "Meltdown" (New Local AI & Mascot) Custom AI Model: Experience "Meltdown"—our custom-trained local AI model. Downloadable directly in-app or via HuggingFace without inflating the initial app size. Onboarding & Tutorial: Get to know your new mascot with an interactive walkthrough designed to introduce you to advanced tool execution and local workflows. Enhanced Performance & Memory Expanded Device Memory: Unlocked extra memory entitlements to handle larger local models and larger context windows with far greater stability. Lower Heat & Reduced Load: Optimized GPU and CPU thermal management to keep your device running cool and fast during long generation sessions. Core Engine Upgrades: Updated the llama.cpp core build, resolving double mmproj bugs while stabilizing Speculative Decoding and Multi-Token Prediction (MTP). Tool Calling & Sub-Agent Upgrades Smarter Execution: Significantly increased execution accuracy and tool usage reliability across all local models. Fixed Tool Routing: Resolved issues where sub-agents failed to operate in edge cases, alongside fixes for harness execution. Crashes & UI Fixes Fixed a crash when clicking the remote connection button from the chat sidebar. Resolved crash issues when rapidly switching tabs. Refreshed layout elements, updated domains, and overall UI polishing. Enjoying iRun? Leave us a review on the App Store!

What’s New in Version 0.2 iRun goes completely mobile! We’ve brought the full power of our privacy-first local AI engine from macOS straight to your iPhone and iPad with incredible portability optimizations and zero quality loss. Here is what's new in this massive update: Full iOS Portability & Optimizations Complete Desktop Features on Mobile: Take your workflow on the go! The AgentView (file explorer, editor panel, and terminal logs), DeepSearchView (web research loops), and Design tab (vector wireframe rendering) are now fully native on iOS. iOS Tool Router: A resilient, built-in fallback router allows iOS to execute web, file, and system notifications completely standalone without needing a Mac host. Buttery-Smooth 120Hz UI: We rebuilt our animation system to leverage Apple ProMotion displays. Chat bubbles, typing indicators, and status updates now glide at a native 120Hz on iPhone 15 Pro and ProMotion iPads. Deep Tech & AI Intelligence Upgrades Speculative Decoding: Experience blazing-fast local generation. The engine now autonomously predicts text tokens ahead to collapse code blocks and text onto your screen instantly. Cross-Platform Network MCP: Connect to remote Model Context Protocol (MCP) servers via Server-Sent Events (SSE/HTTP) or WebSockets across both iOS and macOS. Advanced Subagents: ReAct models can now spawn nested, child subtasks to delegate complex processing steps offline via a structured JSON bridge. Cloud users (Gemini, GPT-4o) can also seamlessly route subagents through their active cloud provider without needing a local model loaded. Creative Workspaces & Customization 2D/3D Design Workspace: Use the AI agent to draft custom 2D canvas drawings, build/animate 3D primitives directly in local code blocks, or edit pixel art using the new 2D Pixel Grid. Dynamic Graphing: Generate clean line, bar, and pie charts instantly from standard text prompts. New Themes & Fonts: Match your aesthetic with the soft pastel Cat Theme or deep navy Ocean Theme. Customize your chat environment with adjustable font sizes and families Bug Fixes & Stability Added a resilient JSON pre-processor to strip AI-generated comments and trailing commas before decoding. Fixed a collapsing layout bug in iOS message bubbles by enforcing strict minimum width constraints. Love using iRun? Please consider leaving us a review on the App Store!