Computes the whole-image SigLIP embedding for anything missing one — images directly, videos by sampling frames (the same approach as the GPU agent). Runs on the ml-worker's CPU, so search, similarity and head suggestions work without a GPU agent; new imports are embedded this way automatically. Detection, cropping and character (CCIP) embeddings are GPU-agent-only. Safe to re-run. To re-embed under a NEW model, use the GPU agent's "Re-embed library" instead.
Turn OFF if you run the GPU agent and removed the ml-worker container — imports then stop queueing CPU embed work nothing will consume (the daily GPU embed backfill covers those images instead).