Back to browse
GitHub Repository

Native macOS semantic search over your local files - text, images, audio, video in one vector space, on-device on Apple silicon.

3 starsSwift

Omni – airgapped macOS multimodal search over local files

by artex_xh·Jun 7, 2026·1 point·0 comments

AI Analysis

●●●BangerWizardryDark HorseSolve My Problem

Multimodal embeddings in one vector space—text queries find images and audio locally.

Strengths
  • MLX-Swift native port runs entirely on-device with no Python or server dependencies.
  • Single vector space unifies text, images, audio, and video for cross-modal search.
  • Airgapped by design—model downloads once, then zero network required for indexing or search.
Weaknesses
  • Apple Silicon and macOS 14+ only excludes Intel Macs and other platforms.
  • Model sizes (2-3GB) may be prohibitive for users with limited storage.
Category
Target Audience

Mac users with privacy concerns, researchers, developers

Similar To

Raycast · Obsidian with embeddings plugins · Spotlight

Similar Projects