RawCull Documentation

• 1: Culling Photos
• 2: Sharpness Scoring
• 3: Similarity and Bursts
• 4: Focus Mask
• 5: AI in RawCull
• 5.1: AI Support in RawCull
• 5.2: CLIP Models and Beta Testing
• 6: RawCull Screenshots
• 6.1: RawCull Screenshots
• 7: Focus Points
• 8: Cache
• 9: Memory Pressure
• 10: Settings
• 11: Security & Privacy

brew tap rsyncOSX/cask && brew install --cask rawcull

1 - Culling Photos

RawCull records decisions without changing the source photos. Add a folder as a catalog, then use Loupe for one-photo review or Grid for a broader overview.

Rate and Navigate

A rating key saves the decision and advances to the next photo. The colored controls in the toolbar filter what is shown; they do not change ratings.

In Grid view, use Command-click to select separate photos or Shift-click to select a range. A rating key applies to the selection. Select two or more photos and choose Compare; RawCull compares up to the first four selected photos.

Inspect a Photo

Double-click a thumbnail to open the full-window viewer. From there you can zoom, rate, show focus aids, and switch between the embedded JPG and a developed RAW preview when supported.

Open the information panel to see the histogram, culling evidence, file and camera details, and quick actions for the selected photo.

Useful viewer keys are +/- for zoom, J for embedded JPG, R for developed RAW, F for focus mask, A for focus point, and Escape to close.

Copy Selected RAW Files

Choose Copy, select a destination, and choose either all rated files or a minimum rating from 2 to 5. Dry run is enabled initially so you can check the result before copying.

RawCull copies files with the system rsync tool. It does not delete source files, and existing newer destination files are not overwritten.

Export JPGs

Select one or more photos and use Actions -> Extract JPGs (Command-J). You can export the embedded JPG or, for supported Sony files, a demosaiced RAW JPEG. Choose a destination folder before starting.

Use Sharpness Scoring and Similarity for large catalogs or burst sequences.

2 - Sharpness Scoring

Sharpness scoring estimates image detail and sorts the strongest candidates first. It is a comparison aid, not an automatic reason to reject a photo.

Score Photos

1. Open Grid view.
2. Choose Score Sharpness.
3. Leave the Sharpness sort enabled to show higher scores first.

RawCull scores the current multi-selection, the active star-rating filter, or the full catalog. It first calibrates the focus threshold to the selected photos, then saves the resulting scores and detected subject labels.

Choose Re-score after changing scoring parameters. Canceling a run discards that run’s results.

Scoring Parameters

For normal culling, use Fast quality with Embedded Preview. Use Balanced or High Precision when small detail matters, and RAW Demosaic only for slower final checks on supported files.

The parameter sheet also controls thumbnail size, border exclusion, subject classification, and how strongly the detected subject affects the score. Larger images and RAW demosaicing take longer.

Good Practice

• Compare scores only within the current catalog and scoring setup.
• Inspect important candidates at high zoom.
• Use Focus Mask to see where RawCull detects detail.
• In a burst, combine sharpness with expression, pose, framing, and timing.

3 - Similarity and Bursts

Similarity groups visually related frames into bursts and suggests the strongest candidates. It is intended to shorten review, while leaving the final choice to you.

RawCull 3 can use the CLIP model selected in Settings > AI. If that model is unavailable, visual grouping falls back to Apple Vision. Semantic search requires a compatible CLIP index.

Analyze a Catalog

1. Open Similarity.
2. Choose Analyze Bursts.
3. Open Needs Review when analysis finishes.

RawCull runs any missing sharpness scoring and similarity indexing automatically. Use Re-index after the catalog changes or when you want to rebuild the analysis.

CLIP indexes are model-specific. Re-index after switching between the OpenAI and DataComp models; embeddings created by one model are never reused by the other.

The similarity slider controls grouping: lower values make tighter groups; higher values include more related frames.

Review Bursts

Each group can include a suggested pick and supporting sharpness or subject information. Open a group to inspect its filmstrip, rate frames, defer the group, mark it reviewed, or open Compare for a closer view. In RawCull 3, Deep Review adds subject-mask, focus, and sharpness evidence to the comparison.

Use Set pick to override the suggestion. One-click Keep Best rates the suggested frame 3 stars and rejects the other frames; Keep Top Two rates the first two candidates 3 and 2 stars and rejects the rest. Review the suggestion before applying either action.

The main queues are:

Analysis results are cached for the catalog. Ratings and manual picks are saved with the rest of the culling data.

Semantic Search in RawCull 3

After CLIP indexing finishes, enter a short English description such as bird in flight or backlit portrait. RawCull ranks the catalog by relative text-to-image similarity. The ranking is not a confidence score, so inspect the results before making culling decisions.

4 - Focus Mask

The Focus Mask highlights areas with strong edge detail. Use it in the full-window viewer or comparison view to check where a photo appears sharp.

Press F or use the focus-mask control. For a more detailed check, switch from the small thumbnail to the embedded JPG (J) or developed RAW preview (R) when supported.

Sharpness scoring calibrates the mask threshold for the current catalog. You can fine-tune it in RawCull -> Settings -> Focus:

The mask is a visual guide. Noise, texture, depth of field, and sharpening in the camera preview can affect the result.

5 - AI in RawCull

RawCull adds optional, locally running AI features to its existing photo-culling workflow. Photographs, prompts, embeddings, masks, and inference results remain on the Mac.

Read:

• AI Support in RawCull for supported features, models, downloads, and system requirements.
• CLIP Models and Beta Testing for comparing the OpenAI and DataComp models.

5.1 - AI Support in RawCull

RawCull AI is planned for release when macOS 27 becomes publicly available. It adds local AI-assisted search and review while retaining the same culling workflow and non-AI functions as the current RawCull version.

The AI features run locally on Apple Silicon. RawCull does not upload photos to an external inference service.

AI-Assisted Culling

RawCull uses two types of vision model:

• CLIP converts images and text into comparable vectors. RawCull uses those vectors for semantic search, visual similarity, and burst grouping.
• SAM 3 locates the subject in an image. Deep Review uses the resulting subject mask together with sharpness and camera autofocus evidence to help compare frames.

AI results are review aids, not automatic decisions. Semantic-search scores describe relative similarity rather than confidence, and the photographer always makes the final selection.

Similarity and CLIP

CLIP does not create captions, keywords, or fixed labels while indexing. It converts each image into a normalized numeric embedding that summarizes its overall visual content. RawCull stores that embedding locally and can reuse it for both similarity analysis and semantic search.

Build the Similarity Index

Choose a CLIP model in Settings > AI, enable it for similarity, and select Index Similarity or Re-index in the burst workspace. Indexing runs the image encoder once for each photograph that does not already have a compatible cached embedding.

CLIP embeddings are specific to the selected model and its preprocessing configuration. Changing the CLIP model or installing an incompatible model version requires a new index. If CLIP is unavailable, RawCull can use the macOS Vision feature-print fallback for visual grouping, but text-based semantic search requires compatible CLIP embeddings.

Group Similar Frames

Select Analyze Bursts after indexing. RawCull compares the cached image embeddings and groups visually related frames for review. The Similarity control adjusts how tightly frames are grouped: a lower value creates tighter groups, while a higher value admits more visually related frames.

The burst list shows each group and highlights its suggested pick. From here you can open the burst, run Deep Review, mark it reviewed, or defer it until later.

Review a Burst

Open a burst to inspect its frames in the filmstrip and compare them at a larger size. The review workspace combines the candidate rank with sharpness, focus-point, saliency, metadata, and subject evidence. You can navigate between frames, assign ratings, pick or reject a photograph, and return to the burst list when the review is complete.

The suggested pick and component scores are starting points, not final judgments. Check expression, pose, timing, framing, and critical focus before deciding which frame to keep.

Semantic Search

After a catalog has been indexed with a compatible CLIP model, enter a short description such as puffin, raven, or squirrel. RawCull ranks the catalog using the cached image embeddings, so later searches do not have to reprocess every image.

Search terms work best in English because the supported CLIP models were primarily trained and evaluated with English text.

Supported Models

RawCull supports one selected CLIP model for similarity and semantic search, plus SAM 3 for subject-aware Deep Review.

The upstream files on Hugging Face are the source models. RawCull requires model bundles converted and validated for Apple Core AI on macOS 27.

Model Downloads

The AI models are not included in the RawCull application or its release download. Open Settings > AI and select Download AI Models to view the available models, their purpose, publisher, version, licence, and installation status.

Downloads use macOS Managed Background Assets. macOS stores and manages each asset pack, and its location can change between app launches. RawCull validates an installed model before enabling it and falls back safely when a required model is missing or invalid.

Review the licence in the model manager before selecting Download. Progress and cancellation are shown in the same window. After installation, the models run locally; photographs are not uploaded as part of downloading or using a model.

Keeping the models separate makes the application download smaller. It does not remove the upstream model’s licence conditions. Follow the RawCull release notes for the model versions supported by each beta or release rather than installing an arbitrary conversion.

Model Licences

The RawCull application licence does not replace or extend the licences for the separately downloaded models.

• OpenAI’s CLIP source repository is published under the MIT License.
• The DataComp model page identifies its licence as MIT. Its model card also documents the training data, intended uses, and limitations.
• SAM 3 is distributed under Meta’s separate SAM License, not the MIT License. Access to the official Hugging Face files may require signing in, sharing the requested contact information, and accepting Meta’s terms.

Review the complete licence and model card shown by RawCull before downloading or using a model. RawCull records the exact model revision and licence version applicable to each converted bundle. If a model update changes its licence, RawCull must present the new terms before downloading that update.

macOS Support

RawCull AI requires macOS 27 and Apple Silicon and will be released when the public version of macOS 27 is available.

At that point, active RawCull development and support move to macOS 27. The existing macOS 26 version will no longer receive support or new features, but installed copies will continue to work with their current functionality.

The macOS 27 AI version retains the current RawCull feature set and workflows. AI support is additive: catalogs, ratings, culling, comparison, sharpness tools, previews, and copying continue to work as they do in the current version.

5.2 - CLIP Models and Beta Testing

RawCull supports two optional CLIP models for semantic search, visual similarity, and burst grouping. This page explains what CLIP does, how the models were evaluated, and how to compare them during the RawCull beta.

Both models run locally on Apple Silicon. RawCull does not upload photographs, search text, or model results to an external inference service.

What CLIP Does

CLIP connects text and images by converting both into numeric embeddings. An embedding represents visual and semantic characteristics in a form that can be compared efficiently.

RawCull uses CLIP in two ways:

• Semantic search: a description such as a dog on a beach, city street at night, or sharp portrait is compared with the indexed photographs.
• Visual similarity and burst grouping: image embeddings are compared with each other to find related frames.

CLIP does not generate captions, modify photographs, rate artistic quality, or make final culling decisions. Its search scores are relative similarity values, not confidence percentages. The photographer remains responsible for every selection and rejection.

The Two Models

The models use different weights, image sizes, tokenization details, and converted Core AI graphs. They therefore do not rank every photograph in the same way. A higher raw score from one model cannot be compared directly with a score from the other.

How the Models Were Evaluated

RawCull uses two independent forms of evaluation.

Conversion parity

The first test compares embeddings produced by each converted Apple Core AI bundle with embeddings from its original source framework. This detects errors in model conversion, tokenization, preprocessing, output selection, and normalization.

Text and image parity are tested separately. A cosine value of 1.0 means that the normalized source and Core AI embeddings are identical. Very small differences are expected from Float16 conversion and differences in image decoding or resizing.

Product behavior

The second test builds complete, model-specific indexes for the same photo catalog and runs the same 77 text queries. It checks result diversity, repeated “hub” images, query time, paraphrase behavior, and image-similarity neighborhoods.

Conversion parity answers whether Core AI reproduces the source model. Product testing answers whether the model is useful for photographers. One test cannot replace the other.

Current Results

The current Core AI bundles produced these results on the fixed conversion fixtures and the 77-query RawCull test:

The two models selected the same first-ranked photograph for only 19 of the 77 queries, or approximately 25 percent. This is the most important beta-testing finding: the models offer meaningfully different search behavior.

OpenAI currently passes RawCull’s chosen 0.998 end-to-end conversion-parity gate. DataComp’s text path is extremely close to its source implementation, but its image path has a larger difference that remains under investigation. That difference does not by itself mean that DataComp produces worse semantic search. DataComp can still retrieve better results for a particular query or catalog, and the initial product test produced slightly more diverse first results.

The test set is not yet a large human-labeled accuracy benchmark. Beta feedback is therefore valuable, especially when it describes which model returned a more useful photograph and why.

Why Beta Testers Should Download Both

Downloading both models lets you compare two genuinely different views of the same catalog. OpenAI may work better for one subject or photographic style, while DataComp may work better for another. Testing both also helps RawCull identify weak query categories and choose sensible model-specific defaults.

Each model is optional. Downloading both uses more storage and creating an index for each takes additional time. A user who does not want to participate in comparison testing can install only one model. For this beta, however, installing both is the most useful configuration.

The models are downloaded through macOS Managed Background Assets. macOS stores and manages the asset packs, and RawCull validates a model before enabling it.

Download Both Models

1. Open RawCull > Settings > AI.
2. Select Download AI Models.
3. Review the licence and model information for each CLIP model.
4. Download OpenAI CLIP and DataComp CLIP.
5. Wait until both downloads report Installed.
6. Close the download window and select Check Again if either model has not yet appeared as available.

Downloaded models run locally. Removing a managed model from the same window removes its RawCull asset pack but does not delete manually installed models.

Compare the Models

Use the same photographs and search phrases for both tests.

Test OpenAI CLIP

1. In Settings > AI, choose OpenAI under Selected CLIP model.
2. Enable Use selected CLIP model for similarity.
3. Open the catalog you want to test.
4. Select Index Similarity or Re-index and wait for indexing to finish.
5. Try a set of short English descriptions and record which results are useful.
6. If you use burst analysis, inspect the groups without changing the source catalog before testing the second model.

Test DataComp CLIP

1. Return to Settings > AI and choose DataComp.
2. Keep Use selected CLIP model for similarity enabled.
3. Re-index the same catalog. Embeddings from OpenAI and DataComp are not interchangeable, so the DataComp test requires its own index.
4. Repeat exactly the same searches and burst-analysis workflow.
5. Compare the photographs, rankings, and groups rather than comparing the raw numeric scores between models.

Useful test phrases include:

• objects and scenes: a dog, mountains surrounding a lake;
• colors and attributes: yellow flower, red car;
• actions and relationships: person riding a bicycle, bird in flight;
• photographic properties: sharp portrait, motion blur, backlit subject;
• paraphrases: try several different descriptions of the same idea.

Search primarily in English during this beta because the supported models were trained and evaluated mainly with English text.

What to Report

Useful feedback includes:

• RawCull beta version, macOS build, and Mac model;
• selected CLIP model;
• approximate number and type of photographs in the catalog;
• the exact search phrase;
• which model returned the more useful first five results;
• examples of clearly relevant or irrelevant results;
• unexpected repeated results across unrelated searches;
• indexing failures, model validation errors, or unusually slow operation; and
• whether burst groups became more or less useful after switching models.

Screenshots are helpful, but do not share private photographs unless you are comfortable doing so. A text description of the expected and actual result is enough.

Choosing a Model After Testing

There is no universal winner yet. Choose the model that gives the most useful results for your photographs and vocabulary. OpenAI is the established reference, while DataComp is a promising alternative with noticeably different retrieval behavior.

You can switch later, but changing models requires a compatible index for the new selection. RawCull keeps model identities separate so that embeddings from one model are never silently reused with the other.

For the broader feature overview, see AI Support in RawCull.

6 - RawCull Screenshots

See how RawCull moves from detailed photo inspection to AI-assisted search and burst review, while keeping the photographer in control of every decision.

Read:

• Explore the RawCull screenshots.

6.1 - RawCull Screenshots

This visual tour shows how RawCull helps you inspect, group, search, and review photos before making the final selections.

Loupe View

The Loupe view keeps the selected photo large while nearby frames remain within reach. The information panel brings together the histogram, sharpness and focus results, file details, and camera settings needed for a careful decision.

Similar Photos in Grid View

Choose a reference photo and RawCull uses CLIP to bring visually related frames together. The grid makes it easy to compare poses and timing across the catalog.

Burst List

RawCull groups related frames into bursts and highlights a suggested pick. You can open a burst, run a deeper review, mark it complete, or defer the decision.

Burst Review

The burst reviewer places every frame in a filmstrip beneath a large preview. Scores and camera details provide evidence, while you choose the strongest moment, rate it, or reject it.

Semantic Search

Describe what you want to find in everyday language. RawCull ranks the locally indexed catalog by meaning, as shown here for the search puffins in flight.

AI Settings

The AI settings show which local models and supporting features are ready. Here you can select the CLIP model used for similarity indexing and semantic search.

Model Downloads

RawCull lists each optional model with its purpose, source, licence, and current status. Models run locally after installation, and photographs are not uploaded as part of the download.

Download progress is shown in the same window, where an active transfer can also be cancelled.

7 - Focus Points

Focus Points show the autofocus location stored in supported Sony ARW and Nikon NEF metadata.

Open a photo in the full-window viewer or comparison view, then press A or use the focus-point control. The marker is shown only when RawCull can decode a valid focus location for that file.

The point shows where the camera reported focus; it does not prove that the subject is sharp. Use it together with Focus Mask and a high-resolution preview.

Support varies by camera model, firmware, and RAW format. If no marker appears, the file may not contain a supported focus location.

8 - Cache

RawCull caches previews so reopening and scrolling through a catalog is faster.

It uses separate memory caches for previews and grid thumbnails, plus disk caches for thumbnails and full-size embedded JPG previews. If an item is not cached, RawCull reads it from the original RAW file and caches the result.

Select Cache JPGs in Loupe view to prepare missing full-size embedded previews for the current catalog. This can make later zooming and comparison more responsive.

Open RawCull -> Settings -> Cache to see current cache use. Clear Disk Cache removes thumbnail files, and Clear JPG Cache removes full-size preview files. Both caches are rebuilt as needed; clearing them does not change source photos, ratings, or exported JPGs.

Memory limits adapt to the Mac’s available unified memory. Under memory pressure, RawCull reduces or clears memory caches automatically. See Memory Pressure.

9 - Memory Pressure

Large catalogs and high-resolution previews can use substantial unified memory. RawCull monitors the macOS memory-pressure level and adjusts its caches automatically.

When pressure returns to normal, RawCull recalculates its normal cache limits. Source photos and saved ratings are not affected.

The Memory settings tab shows total and used memory, RawCull’s memory use, and the current system pressure. For troubleshooting, Diagnostics -> Memory Console records detailed cache and memory samples.

If warnings continue, close other memory-heavy apps, stop the current task with Actions -> Abort task (Command-K), or work with a smaller catalog.

10 - Settings

Open RawCull -> Settings. The current release provides four settings tabs; RawCull 3 adds an AI tab on macOS 27.

Sharpen Zoom Preview develops the RAW through macOS and applies micro-detail sharpening. It is slower than using the embedded JPG and may not be available for every file.

Use Save Settings after changing thumbnail or focus values. Reset to Defaults restores the values in that settings area. Scoring options are available from Scoring Parameters in the main window.

In RawCull 3, Settings > AI reports which CLIP and subject-review resources are available. Select either OpenAI or DataComp CLIP and enable it for similarity before indexing a catalog. Download AI Models opens the model manager, while Check Again refreshes availability after an installation or removal.

11 - Security & Privacy

RawCull is designed for local photo culling. Core culling works offline, and RawCull 3 performs AI indexing, search, and review on the Mac.

File Access

RawCull runs in the macOS App Sandbox. It can read a catalog or write to a destination only after you choose that folder. macOS security-scoped bookmarks allow previously approved folders to be used again.

Copying is non-destructive: RawCull copies the chosen RAW files with the system rsync tool and does not delete files from the source catalog.

Local Data

RawCull stores settings, approved folder locations, ratings, sharpness results, burst choices, and rebuildable preview caches on your Mac. RawCull 3 also stores embeddings, masks, and AI review results locally. Caches can be cleared from Settings.

RawCull does not use analytics, telemetry, cloud inference, cloud sync, advertising, or tracking. Photographs, search descriptions, embeddings, masks, and inference results are not sent to an external AI service.

Optional AI model downloads use macOS Managed Background Assets and therefore require a network connection. macOS stores and manages those model resources; after installation, RawCull runs them locally. Model downloading does not upload photographs.

RawCull’s privacy manifest declares no tracking and lists only the required system API access reasons.

RawCull does not request access to the Photos library, camera, microphone, location, contacts, calendars, Full Disk Access, iCloud, Bluetooth, screen recording, or accessibility services.