Google
EmbeddingGemma 2 puts video search on a phone at 740M parameters
Google's open model unifies text, images, audio and video in one vector space, so a voice memo can find the matching shot without leaving the phone.
Every SLOP TV News story about on-device.
Google's open model unifies text, images, audio and video in one vector space, so a voice memo can find the matching shot without leaving the phone.