Feature Extractors

Detect and identify faces in images with high accuracy

Generate vector embeddings for video content

Extract structured data from webpages while maintaining semantic context and relationships

Extract semantic embeddings from documents, transcripts and text content

Generate visual embeddings for similarity search and clustering

Extract semantic embeddings from audio content for similarity search

Unified embeddings for video, audio, image, and text — scene/silence chunking, Whisper transcription, thumbnails, and Gemini vision.

All-in-one extractor for image, video, audio, and documents — auto-detects modality and applies the right pipeline.

Embed all files of an object (images, PDFs, video, audio, text) into a single 3072-D Gemini vector.

Decompose PDFs into spatial blocks — paragraphs, tables, forms, headers — with layout classification and E5 text embeddings.

Store and canonicalize objects with zero ML — metadata-only ingestion for bucket/object modeling without embeddings.

Read scrolling/marquee video text via phase-correlation band detection, panoramic stitching, and VLM OCR.