We present omni-macos, a search engine that embeds text, code, documents, images, audio and video into one representation space and runs its encoder, index and store on the Mac that already holds the files, so the files, the queries and the vectors stay on that machine. It keeps a background indexer and an interactive search box responsive on one shared device. Each distinct chunk is embedded and stored once, indexer work is split into smaller units while the user types, queries scan a one-bit replica of the index and rescore a shortlist exactly, and one memory setting bounds every allocator that draws on unified memory. We measure omni-macos on five Macs spanning an eightfold range of accelerator width and a thirty-twofold range of memory. omni-macos is open source under Apache 2.0 at https://github.com/hanxiao/omni-macos.
翻译:暂无翻译