Local AI Culling

Free plan LinuxMacWindows

What Local AI Culling is

Local AI Culling is an open-source application for professional photographers who want AI-assisted photo selection to run offline. It analyzes a shoot and sorts images into KEEP, REVIEW, and REJECT folders for later editing in Lightroom or another editor. It groups visually identical photographs taken moments apart, detects high-speed bursts, and ranks the frames in each burst. Its quality scoring considers focus, noise, composition, facial expressions, and editability; it also supports subject detection, blur detection, closed-eye detection, and RAW photos.

A local web dashboard lets photographers inspect decisions, change thresholds, and see why a result was reached. By default, export uses hardlinks to organize files without copying or changing the originals, with copy mode available as an alternative. The project says images are not uploaded, cloud APIs are not used for image or metadata processing, and usage telemetry is not collected or sent. It runs on Windows, macOS, and Linux, with acceleration varying by system. Installation requires Python 3.10 or later; Python 3.11 is recommended. Supported formats include JPG, JPEG, PNG, CR2, and NEF. The project recommends 500–2,000 images per run; runs above 5,000 may increase memory use and slow duplicate processing.

Who it suits

Local AI Culling suits professional photographers who want to sort shoots locally before editing in Lightroom or another editor. It is a fit for photographers who need duplicate grouping, burst ranking, and technical image scoring across Windows, macOS, or Linux.

What it does well

Runs offline on Windows, macOS, and Linux.
Groups visually identical images and detects high-speed bursts.
Scores focus, noise, composition, expressions, and editability.
Local dashboard shows decisions and allows threshold changes.
Default hardlink export leaves originals unmodified and unduplicated.
Open source under the MIT License.

What to know first

Requires Python 3.10 or later to install.
CR3, compressed ARW, and HEIC support may be incomplete.
Lightroom Classic plugin integration and video culling are postponed.
Does not edit Lightroom catalogs or generate XMP sidecars.

Limits

The project recommends 500–2,000 images per run and warns that datasets above 5,000 may cause memory pressure and slower duplicate processing. Some formats, including CR3, compressed ARW, and HEIC, may have incomplete support or metadata extraction.

Getting started

Open the project website on GitHub.
Install Python 3.10 or later; Python 3.11 is recommended.
Use an SSD, which the project highly recommends.
Run a batch of 500–2,000 images for the recommended dataset size.
Review the KEEP, REVIEW, and REJECT results in the local web dashboard.
Export using hardlinks by default or configure copy mode.

Verdict

Choose Local AI Culling if you are a professional photographer seeking offline culling, burst ranking, and quality scoring without subscription fees. Look elsewhere if you need Lightroom Classic plugin integration, catalog editing, XMP sidecars, or video culling.

Questions

Is Local AI Culling free?

Yes. It is free, has a free plan, and is open source under the MIT License.

Which operating systems does it support?

The README lists Windows, macOS, and Linux. Acceleration depends on the platform and may use CUDA, Metal/MPS, or CPU.

Does it support RAW photos?

Documented supported RAW formats include standard CR2 and NEF. CR3 and compressed ARW may have incomplete support.

Does it integrate with Lightroom?

Organized images can be edited in Lightroom or another preferred editor. Lightroom Classic plugin integration is postponed, and the app does not edit catalogs or generate XMP sidecars.

Are images uploaded for processing?

The security policy says images are not uploaded and cloud APIs are not used for image processing or metadata analysis. It also says usage telemetry is not collected or transmitted.

How many images should I process at once?

The project recommends 500 to 2,000 images per run. Datasets over 5,000 may use more memory and slow duplicate processing.

Facts from Local AI Culling's own pages

Burst ranking: It identifies high-speed bursts and selects the best frame in each burst. · github.com, 30 Sept 2026
Culling features: It detects duplicates and bursts, ranks burst frames, and assesses focus, noise, composition, and facial expressions. · github.com, 30 Sept 2026
Dataset limits: The project recommends 500–2,000 images per run and warns that datasets over 5,000 images may use substantially more memory and process duplicates more slowly. · github.com, 30 Sept 2026
Dataset size: The project recommends 500 to 2,000 images per run and warns that datasets over 5,000 images may cause memory pressure and slower duplicate processing. · github.com, 30 Sept 2026
Duplicate detection: It groups visually identical photos taken moments apart. · github.com, 30 Sept 2026
Editor workflow: The README says organized images can be edited in Adobe Lightroom or a preferred editor, while the limitations page says Lightroom Classic plugin integration is postponed. · github.com, 30 Sept 2026
Export: The default hardlink export organizes files without duplicating or modifying the originals; copy mode is also configurable. · github.com, 30 Sept 2026
Feedback dashboard: A local web dashboard lets photographers review decisions, adjust thresholds, and see why the AI made a decision. · github.com, 30 Sept 2026
Image evaluation: Its scoring evaluates technical quality including focus, noise, composition, expressions, and editability. · github.com, 30 Sept 2026
Image-format limits: Supported formats include JPG, JPEG, PNG, CR2, and NEF; CR3, compressed ARW, and HEIC may have incomplete support or metadata extraction. · github.com, 30 Sept 2026
Image formats: Documented supported formats include JPG, JPEG, PNG, and standard CR2 and NEF RAW; CR3, compressed ARW, and HEIC may have incomplete support. · github.com, 30 Sept 2026
Installation: Installation requires Python 3.10 or later, with Python 3.11 recommended, and an SSD is highly recommended. · github.com, 30 Sept 2026
Integration limits: The documented v1.0 limitations say Lightroom Classic plugin integration and video culling are postponed, and the app does not edit Lightroom catalogs or generate XMP sidecars. · github.com, 30 Sept 2026
Intended users: The application is built specifically for professional photographers. · github.com, 30 Sept 2026
License and fees: The repository identifies the project as open source under the MIT License and says local operation eliminates subscription fees. · github.com, 30 Sept 2026
Other limitations: The project says it does not edit Lightroom catalogs or generate XMP ratings, and video culling is postponed. · github.com, 30 Sept 2026
Platforms: The README lists Windows, macOS, and Linux support, with hardware acceleration where available and CPU fallback. · github.com, 30 Sept 2026
Privacy: The security policy says images are not uploaded, cloud APIs are not used for image processing or metadata analysis, and usage telemetry is not collected or transmitted. · github.com, 30 Sept 2026
Purpose: Local AI Culling is an open-source, offline AI-assisted image culling application built for professional photographers. · github.com, 30 Sept 2026
Support: The project directs users to its documentation and GitHub Issues for help, bug reports, and feature requests. · github.com, 30 Sept 2026
Supported systems: The README lists Windows, macOS, and Linux support, with platform-dependent CUDA, Metal/MPS, or CPU acceleration. · github.com, 30 Sept 2026
Workflow: It analyzes photo shoots and organizes images into KEEP, REVIEW, and REJECT folders for editing in Lightroom or another editor. · github.com, 30 Sept 2026

Where Local AI Culling stands on Freedom251

AI Photo Culling Software#12 of 29