
EmbeddingGemma 2: search beyond text in Flow-Like
EmbeddingGemma 2 is already live in Flow-Like. Build local search across text, images, audio, and video with one shared embedding space.
Tag
5 posts in this topic.

EmbeddingGemma 2 is already live in Flow-Like. Build local search across text, images, audio, and video with one shared embedding space.

A quick comparison of OpenAI dots, Meta Muse and Grok Bot, plus a practical way to build a recurring AI assistant on your own device with Flow-Like.

FlowPilot goes global, flows become editable code, and rich apps gain voice, video, planning, local AI, permissions, Suites, Data Studio, and a faster foundation.

Flow-Like adds local SCRFD face detection, ArcFace embeddings, age and gender estimates, verified model caching, and one ONNX acceleration policy across platforms.

Speech-to-text and text-to-speech nodes land in Flow-Like: hosted providers and fully local, on-device models (Whisper, OLMoASR, Kokoro and more) running on candle — private by default.