Media
5 items
Description
Right-click any image to analyze it with on-device AI. Your images never leave your browser. No hard character cap in practice (limit is very high), but scannable beats exhaustive. Structure: hook → what it does → why it's different → how it works → privacy → footer. ``` See what's hiding in every image — without it ever leaving your browser. Lens adds one simple action to Chrome: right-click any image on the web and choose "Analyze with Lens." A small AI model — running entirely on your device — reads the image and shows you a description in a clean, unobtrusive card right where you clicked. No new tab, no dashboard, no account. WHY LENS IS DIFFERENT Most "AI" browser tools quietly send your data to a server somewhere. Lens doesn't have one. The model runs locally, in your browser, using on-device machine learning (Transformers.js). Your images are never uploaded, logged, or seen by anyone — including us. HOW IT WORKS 1. Right-click any image → "Analyze with Lens" 2. A small on-device vision model describes what's in it 3. Copy a shareable, watermarked snapshot in one click — or share it directly through your device's native share sheet WHAT YOU GET • Instant image descriptions, right where you're already browsing • Automatic tagging (photograph, illustration, screenshot, and more) • A local cache, so re-checking the same image is instant the second time • A one-click "Copy & Share" snapshot, formatted for X, Slack, or Discord PRIVACY, BY ARCHITECTURE Lens has no backend server. There is nothing to send your images to, because nothing exists to receive them. The only network activity Lens ever makes is a one-time download of the AI model's files from Hugging Face's public model hub — not your data going out, just the model coming in. Everything after that runs offline, on your machine. See the full privacy policy: https://wushu75.github.io/Lens/privacy.html