Open-Source SCM: Local AI Search for Every Video Frame and Photo on Mac

Developer Allan Lee (allenv0) published an open-source project under the MIT license, SCM (Screen Memories), on Show HN — a local AI search for all photos and every video frame on macOS without cloud or accounts, with installation via Homebrew and offline operation after a one-time download of model weights.

Compiled by Sergey KostenchukPublished 2026-10-04Updated 2026-10-04

2026-10-04 Coding Meta

🛠 Open-Source SCM: Local AI Search for Every Video Frame and Photo on Mac

Developer Allan Lee (allenv0) released the SCM project under the MIT license: AI search for all photos and videos on macOS without cloud or accounts. Modes: semantic (CLIP, SigLIP), scenes in video with refinement to the timecode, OCR via Tesseract, dialogue via Whisper, plus local chat with Qwen3 and Llama 3.2. Weights are downloaded once, and everything works offline thereafter.

🌍 Private semantic search for a media archive is built entirely from open components on a consumer Mac — without cloud APIs or subscriptions. This is an open-source equivalent of Apple Photos search with precision down to the frame and dialogue.

👤 Installed via Homebrew on Apple Silicon: search for photos and videos by frame description, jump to the desired shot or dialogue, and search for text in images. Nothing leaves the Mac, but indexing is not fast — about 1.5 hours of CPU time for 10,000 photos.

Source 1: https://github.com/allenv0/SCM Source 2: https://news.ycombinator.com/item?id=49952111

Sources