Every frame graded for focus, noise, exposure and composition, with a PASS, WARN or FAIL and a one-line critique. Blur, closed eyes, smeared faces, bad hands, misspelled signage and near-duplicates get flagged before delivery.
The two workspaces do different jobs and are meant to be used in order.
Fast triage of a whole shoot into tiers, with burst clusters grouped
Deep inspection of the keepers, with a deliverable verdict per frame
Thousands of frames become a short, checked, editor-ready set
The lens sets what a failure means. Agency rules only apply when you ask for them.
Expressions, eyes open, coverage
Peak action, sharp subject eye, clean background
Intentional shallow depth of field is never called soft focus
Even light, isolation, straight verticals
Agency megapixel minimums and market saturation
Neutral grading for mixed shoots
Focus (sharpness via Laplacian variance), noise, exposure, composition and effective megapixels, plus face integrity, anatomy and any text in frame. It also clusters near-duplicates with a perceptual hash so you only keep the best frame of each cluster.
No. You pick a lens per batch — wedding, event, portrait, sport, landscape, wildlife, travel, architecture, product, stock, editorial or general — and the grading follows. Stock-specific gates such as market saturation and agency megapixel minimums only switch on under a stock or editorial lens; a wedding batch is never rejected for being 'oversupplied'.
Cull is the fast first pass: it grades a whole shoot and sorts it into tiers so you can drop the misses. Quality Checker is the delivery gate: it inspects the frames you kept in more depth — faces, anatomy, text, resolution — and gives each one a PASS, WARN or FAIL with a written critique. Most people cull first, then run QC on the keepers.
Lightroom shows you everything and asks you to flag it. The Quality Checker reads every frame, assigns a verdict with a short critique, and groups visual duplicates, so the keep-vs-cull decision is one click instead of one hour.
Yes. Face integrity checks for melted or smeared features, anatomy checks for wrong hands and limbs, and a text check catches misspelled signage — the failures a fast human pass usually misses at thumbnail size.
Yes — Canon CR2/CR3, Nikon NEF, Sony ARW, Adobe DNG, Fujifilm RAF, Olympus ORF and Hasselblad 3FR. The embedded preview is extracted in your browser; only a 1280px JPEG preview is uploaded.
Every gate is a switch: megapixels, market saturation, composition, anatomy, text and face integrity. Turn one off and nothing it would have flagged counts against a frame — including in the reasoning shown on rejects.
Quality Checker runs at 1 credit per image on every plan. Re-runs cost 1 credit per image; cached passes don't.
Pick a folder and originals stay on your machine — only a 1280px JPEG preview is uploaded for scoring. The vault stores that preview plus the verdict and metadata.
The XMP export zip includes a README plus embed-metadata.sh (macOS/Linux) and embed-metadata.bat (Windows). Install ExifTool (free) and run the script — it copies title, description, keywords, star rating and colour label into each RAW or JPG. Capture One, darktable, Photo Mechanic and DigiKam also read the .xmp sidecars natively, no script needed.
Free tier included. No card required, nothing to install.
[ LAUNCH STUDIO ]