Linkty vs Gadeon

Links pile up everywhere — in chats, browser tabs, tweets, and notes. You save them "for later," but later never comes. Linkty fixes this. SWIPE TO DECIDE Linkty turns link sorting into a fast, satisfying experience. Each link appears as a card — swipe right to keep, left to skip. One decision at a time, zero overwhelm. TAG AS YOU SAVE Assign tags the moment you save a link. Create new tags on the fly — no need to set up categories in advance. Group your tags for even more structure. YOUR PERSONAL LINK LIBRARY Every saved link lives in a searchable, filterable collection. Sort by date, domain, or tags. Find exactly what you need in seconds. RATE, NOTE, REVIEW Go beyond saving. Rate links from 1 to 5 stars, add personal notes, and mark them as reviewed. Build a curated knowledge base — not just another bookmarks folder. SMART CLIPBOARD DETECTION Copy a link anywhere on your phone — Linkty catches it and suggests adding it to your queue. No switching apps, no manual pasting. BUILT FOR FOCUS No algorithmic feeds. No social features. No distractions. Just your links, your tags, your pace. KEY FEATURES • Swipe-based link triage — keep or skip in a second • Flexible tagging with groups • 1-5 star rating system • Personal notes on every link • Reviewed/unreviewed tracking • Search, filter, and sort your library • Clipboard link detection • In-app browser with Safari fallback • Share extension — save from any app • Dark mode support Linkty is free and works entirely on your device. Your links stay private.

Gadeon runs Qwen3.5 and QwenPaw hybrid language models entirely on your device's Apple Neural Engine, using Core ML as its only framework dependency. There's no C, no Python, and no cloud calls: every token is generated locally, keeping your conversations private. Under the hood, these models mix Gated DeltaNet recurrent layers with a smaller number of attention layers. Gadeon compiles that hybrid architecture directly onto the Neural Engine's dataflow accelerator, so both prompt processing (prefill) and token generation (decode) run on-chip instead of falling back to the CPU or GPU. On an Apple Silicon Mac, prefill measures roughly 2 to 2.4 times faster and about 4.4 times more energy-efficient per token than a Metal GPU baseline, while leaving the GPU free for other work. Conversations in Gadeon are append-only: each turn extends the model's fixed-size recurrent state instead of re-processing the entire chat history, so response latency doesn't grow as a conversation gets longer. The app can also roll back to an earlier point in a conversation, dropping reasoning traces or replaying tool calls in a clean form without reprocessing everything from scratch. For ternary-quantized models, Gadeon switches to a pure-Swift Metal GPU backend behind the same session and tokenizer interface, so both compute paths feel identical from the chat UI. Gadeon is a proof of concept for bringing Gated DeltaNet hybrid models to the Neural Engine through Core ML - built to explore what's possible on Apple's own silicon.