AI tools for your Lightroom Classic library

lrc-ai-assistant has been replaced by LrGeniusAI — a smart Lightroom Classic plugin for AI-powered tagging, describing, and semantic image search.

LrGeniusAI brings modern Large Language Models (LLMs) into Adobe Lightroom Classic: it analyzes your photos, generates accurate tags and detailed descriptions, and lets you search your library with natural language. You can run local models for maximum privacy or use cloud APIs; the plugin fits into your workflow either way.

Core features

• AI Edit & style training (beta) — Create Develop recipes from your own saved edits, with at least five training examples and per-photo review enabled by default. This workflow interpolates your examples rather than calling a language model.

• Image culling (beta) — Group bursts and near-duplicates, rank frames, and create collections for picks, alternates and reject candidates.

• AI-powered tagging & describing — LLMs recognize image content and generate metadata and detailed descriptions.
• Semantic free-text search — Find images by describing what you want (e.g. “Red sports car parked in front of a garage”). LrGeniusAI builds a relevance-sorted Collection in Lightroom from your prompt.
• Built-in local AI — The backend runs vision models itself: MLX on macOS (Apple silicon) and llama.cpp on Windows. Pick a model in the Plug-In Manager and click Download — no external app needed.
• Local & cloud models — Also works with Ollama and LM Studio, or cloud providers Google Gemini and ChatGPT/OpenAI.
• People & faces — Detect and cluster faces, name persons, browse person collections, and find similar faces across the catalog.
• Species identification (on-device) — Identify animals, plants and fungi down to the species with BioCLIP 2, running entirely on your machine.
• Customizable prompts & temperature — Edit system prompts and control creativity vs consistency in the Plug-In Manager.
• Photo context — Add hints (names, background details) in a dialog or in Lightroom’s metadata panel so the AI can use them.
• Custom Rust backend — A local server (geniusai-server) written in Rust for low memory overhead. Import existing catalog metadata before the first AI run.

Tech stack

• Plugin: Lua (Lightroom Classic SDK)
• Backend: Rust — geniusai-server, an axum HTTP service that runs locally alongside Lightroom
• Embeddings & semantic search: SigLIP2 via ONNX Runtime
• Faces: YuNet (detection) + FaceNet (embeddings), ONNX
• Species: BioCLIP 2 (ONNX) with a pruned TreeOfLife taxonomy head
• Local inference: MLX on macOS (Apple silicon, Metal helper process); llama.cpp compiled into the backend on Windows (GGUF, Vulkan)
• Database: LanceDB
• Supported model providers: built-in MLX (macOS), built-in llama.cpp (Windows), Google Gemini, ChatGPT/OpenAI, Ollama, LM Studio
• License: AGPL-3.0

The backend used to be a Python/Flask server; it was rewritten in Rust and now lives in the LrGeniusAI repository under server-rs/. The old geniusai-server repository is archived. Google Vertex AI support was removed in August 2026.

Get it

• Download: GitHub Releases
• Website & help: lrgenius.com · lrgenius.com/help
• Source: github.com/LrGenius (LrGeniusAI, LrGeniusTagAI)

Frequently asked questions

What is LrGeniusAI?

LrGeniusAI is a Lightroom Classic plugin that uses Large Language Models (LLMs) to automatically tag and describe your photos, and lets you search your library with natural-language prompts. It ships built-in local AI (MLX on macOS, llama.cpp on Windows), and also supports Ollama, LM Studio, Google Gemini and ChatGPT/OpenAI.

Does LrGeniusAI work with local AI models?

Yes. LrGeniusAI runs vision models itself — MLX on macOS (Apple silicon) and llama.cpp on Windows — so no external app is required. It also supports Ollama and LM Studio, or cloud models such as Google Gemini and ChatGPT/OpenAI. Switch between local and cloud as needed.

How does the semantic search in LrGeniusAI work?

LrGeniusAI builds a vector index of your photos using SigLIP2 embeddings, stored in a local LanceDB database. Enter a natural-language description (e.g. ‘red sports car parked in front of a garage’) and the plugin returns a relevance-sorted Collection in Lightroom.

Where can I download LrGeniusAI?

Download LrGeniusAI from lrgenius.com or from the GitHub Releases page at github.com/LrGenius/LrGeniusAI/releases. Documentation and help are at lrgenius.com/help.