Search models

Search is powered by a model that runs entirely on your machine. Weights are downloaded on first use of a command that needs them, never at install.

Command	Model	Downloaded size
videre embed, videre search	SigLIP, default google/siglip-base-patch16-224	about 780 MB
videre faces	InsightFace WePrompt/buffalo_l (SCRFD detector + ArcFace embedder)	about 180 MB

Both live in the shared Hugging Face cache at ~/.cache/huggingface/hub/, overridable with HF_HOME. See what gets downloaded.

scan, dedupe, fix-dates, prune, stats, locations, and report without similarity search need no model at all.

Choosing one

embed, search, classify, report and mcp all take --model <id>, resolved as --model first, then default_model in your config, then the built-in default.

videre config set model google/siglip2-base-patch16-384   # lasting default
videre embed --model google/siglip2-base-patch16-384      # just this once
videre config                                             # show what resolves

Non-default models are larger and are only fetched if you actually select one: siglip2-base-patch16-384 is about 1.4 GB, and siglip-so400m-patch14-384 about 3.3 GB.

Using more than one

Each model keeps its own data, so they never overwrite each other and you can compare them on the same library:

videre embed                                                   # the default model
videre embed --model google/siglip2-base-patch16-384           # a second one
videre stats                                                   # see what each has
videre search "sunset" --model google/siglip2-base-patch16-384 # search a specific one
videre config set model google/siglip2-base-patch16-384        # make it the default

Preparing a second model does not disturb the first, and switching back and forth invalidates nothing.

Asking for a model you have not prepared gives an error listing the ones you do have, rather than silently returning no results. videre report is the exception: a missing model disables its in-page similarity search with a note, rather than failing a report that works fine without it.

Only videre embed creates a model’s data. Everything else reads.

Where the data is kept

Not in the main database. Each library and model pair gets its own file:

~/.videre/embeddings/<library>-<hash>/<owner>--<model>.db

Per library rather than one shared file per model, because videre prune cannot see another library’s contents. A shared layout would let one library’s cleanup delete data another still needs, and an embedding costs hours to rebuild.

The library part of the path includes a hash of its canonicalised path, so two libraries both called photos.db in different folders never collide.

Expect roughly 130 MB to 190 MB per model for a 70,000 photo library. videre stats reports the actual figure per model.

Upgrading from before 0.10

Data written by 0.9.x lived in the main database. That fallback was removed in 0.11.0, so such a library now reports the same clear error as any other missing model.

Nothing is deleted: the old data sits untouched and can be dropped by hand. Rerun videre embed to rebuild in the current layout.