imagic - AI photo culling + RAW editing you buy once, not rent
by•
imagic culls big shoots with AI (sharpness, closed eyes, duplicates and bursts) and does your batch RAW edits and export, in one desktop app that runs on your own machine. The twist: you buy it once (Lite EUR 19, Plus EUR 49, Max EUR 99), not a subscription. Aftershoot, FilterPixel and Lightroom are monthly; imagic you buy once and own. Claude, Cursor or Codex can drive it via the built-in MCP server. Free trial: 500 exports, no card. Windows and macOS.

p/htmldrag
p/runevr
p/producthunt
p/google-cloud-platform
p/kodlo
p/ai-context-flow
p/jot-2-0
p/beamfy
p/refs
p/dex-by-exmergo
p/reso
p/duckfightclub
p/seshka
p/talanoa-2
p/nudgeforme
p/dino-kids
p/nexupira
p/many-displays
p/murror
p/operum
Replies
Hi PH! I'm Luke, maker of imagic.
I got tired of two things: paying Adobe every month, and paying another monthly subscription (Aftershoot/FilterPixel) just to cull my own photos. So I built imagic - a desktop app that does AI culling AND batch RAW editing in one place, runs on your own machine, and you buy once (from EUR 19) instead of paying $15/month forever.
It scans your RAW folders, scores every shot for sharpness, exposure, closed eyes and detail, groups bursts and duplicates, keeps your best frames, then batch-edits and exports in the background. Windows and Mac. Free trial: 500 exports, no card.
Update, October 2026: it now has a built-in MCP server, so Claude, Cursor or Codex can run a whole cull and edit for you. I'd genuinely love your honest feedback, especially from high-volume shooters. What would make this a no-brainer switch for you?
Does the AI culling actually learn from your accept/reject choices over time, or is it purely rule-based on sharpness and exposure metrics?
@grkem86fq Good question, and it's a bit of both. Out of the box, every frame gets a score from sharpness, exposure, detail, noise, composition, faces and whether the subject is actually in focus, and the card shows you each of those so you can see why it made the call.
On top of that it does learn from you. Every time you flip one of its keep/trash decisions, that correction is saved locally, and once there are a few of them it re-weights those factors and shifts the keep/trash cut-offs toward what you actually keep. Say you keep a lot of slightly soft frames: sharpness stops dragging them into the trash. It keeps a separate profile for people and for action/sports, since the rules differ, and the duplicates view learns which frame of a burst you tend to pick. It all lives on your machine, nothing is uploaded.
It's a gentle, capped adjustment rather than retraining a neural network, so a handful of odd clicks can't wreck it, and it kicks in on your next analysis or re-cut.
Luke