How to Convert a Lightroom Preset to a LUT

A preset is a list of instructions: exposure up this much, this tone curve, these HSL shifts. A LUT is a table of finished answers: this input colour becomes that output colour, repeated some thirty-five thousand times.

Turning the first into the second means evaluating the instructions against every colour and writing down the results. That is straightforward — but it needs a colour engine that can render the preset, and Lightroom does not expose one to export through. Camera Raw's processing is proprietary, versioned, and tied to camera profiles that live outside the sidecar.

So the conversion has to happen somewhere that can genuinely render the recipe. That is what a converter does: it builds an identity cube, renders it through the preset's colour transform, and reads the result back as a lookup table.

The dividing line is not about effort or quality. It is that a 3D LUT is handed one colour at a time and told nothing about where in the frame that colour came from.

Everything global therefore converts exactly: exposure, contrast, highlights, shadows, whites and blacks; white balance and tint; the master and per-channel tone curves; all eight HSL bands; split toning and the colour grading wheels; vibrance and saturation.

Everything spatial cannot convert at all, because there is nowhere in a LUT to put it. That means brush, radial and linear masks, vignettes, grain, texture, clarity, sharpening, noise reduction, lens corrections and crops. A preset that leans on a graduated filter to darken a sky will produce a LUT that does not darken the sky, and no converter can change that. Dehaze is the one that surprises people by converting: despite sitting in the same panel as texture and clarity, it is arithmetic on the colour itself rather than a comparison with neighbouring pixels, so it bakes exactly.

• ·Converts exactly — exposure, contrast, tone curves, HSL, split toning, colour grading, white balance
• ·Cannot convert — masks, vignettes, grain, texture, clarity, sharpening, lens corrections
• ·The honest test: if the effect depends on *where* a pixel is, it will not be in the LUT

Find the .xmp first. Lightroom Classic stores develop presets in Camera Raw's settings folder — on Windows %APPDATA%\Adobe\CameraRaw\Settings, on macOS ~/Library/Application Support/Adobe/CameraRaw/Settings. You can also right-click a preset in the Develop panel and choose Export.

If what you find ends in .lrtemplate rather than .xmp, it is the format Adobe retired in 2018 and no current converter will read it. It has to be brought across to .xmp first.

Then drop the .xmp into RawCut's converter. It parses the sidecar, lists every setting it found so you can confirm the right preset was read, warns you if the preset contains masks, and exports a 33×33×33 .cube. Nothing is uploaded — the parsing and the baking both run in your browser.

The LUT will match the converter's rendering of the preset exactly, because it is baked by rendering through the same engine that draws the preview you are looking at.

Against Lightroom itself, expect close but not identical. Adobe's process versions and camera profiles are proprietary and are not part of the sidecar, so a converter is working from the instructions without access to the whole of Adobe's interpretation of them. For creative looks the difference is usually invisible; for a preset that depends heavily on a specific camera profile it can be noticeable.

One practical consequence worth planning for: a look built on well-exposed stills assumes well-exposed input. Applied to log footage it will come out flat and washed, not because the conversion failed but because the LUT is being fed numbers it was never designed for. Convert the footage to the space the look expects first.

This is the most common complaint after a successful conversion, and it is usually not the conversion's fault. A preset applied in Lightroom sits on top of Adobe's default rendering; the same look applied to already-graded footage is doing contrast work twice.

Grade to a neutral, correct starting point first and apply the look after. If it is still heavy, reduce the LUT itself rather than fighting it — blending a LUT toward the identity transform and re-exporting bakes the reduced strength into the file, which means it survives into software that has no opacity control at all.