Marengo 3.0

Name: Marengo 3.0
Author: Twelve Labs

Twelve Labs 📐 Embedding

Twelve Labs multimodal video embedding model. Converts video, audio, and text into a shared vector space for semantic search across 36 languages.

Specifications

ModalitiesInput: video, audio, text, image · Output: embedding

FeaturesTools: No Streaming: No

Capabilitiesembedding, search

$0.001 / generation

Platform credit pricing — live rates on the Models page.