About PicForge

Why PicForge exists

Everyday image chores are split between upload-based tool sites that collect your files and heavy desktop software that is overkill for a resize. People handling sensitive images — IDs, legal documents, medical photos, family pictures — deserve a tool that never asks them to trust a server.

PicForge answers that with one fast web app where the file never leaves the device. Deterministic processing happens locally; the optional AI features are explicitly opt-in and clearly labeled.

How it is built

PicForge is a static site built with Next.js and React, with all image work running in the browser through Canvas, WebAssembly and background workers. There is no backend that stores user data: the only server-side component is an optional AI endpoint for users who consent to it.

Heavy decoders (HEIC, TIFF, PSD, JXL), on-device AI models and export libraries load on demand, so pages stay fast. The app is installable as a PWA and keeps working offline for the tools you have used.

• 18 tools covering the everyday 90%: resize, compress, crop, rotate, convert, social sizes, favicon, ID photo, EXIF, filters, watermark, photo editor, viewer, blur/censor, color picker, GIF frames, merge PDF and background removal.
• Modern formats in and out, including HEIC, TIFF, PSD and JXL input.
• Batch processing with zip downloads for common workflows.

Principles

• Local-first: the default path never transmits your files.
• No dark patterns: no watermarks, forced sign-ups, artificial waits or upsells.
• Fast by default: tiny initial payloads, compute off the main thread, lazy-loaded codecs.
• Accessible: keyboard-operable, screen-reader friendly, WCAG 2.1 AA targets.
• Transparent: the analytics are cookieless, aggregate and you can turn them off.

Get started

Head to the home page to browse all tools, or read the Privacy Policy to see exactly what the app does — and does not — collect.