Pass IndexThe State of AISign in

Multimodal Embeddings

The multimodalembedding@001 model converts image, video, and text into vectors of floating point numbers, for uses such as image classification, image search, and recommendations.

text + image + video → embedding · made by Google

$0.0001per image
Google Vertex AI

Sold by 1 way

SellerLaneRate
Google Vertex AI 0 ways$0.0001
Google Vertex AI cloudnot on the seller's list todaystandard$0.0001per imagecloud.google.com · read 2026-08-24

About

Multimodal Embeddings — a vectors model from Google.

It takes text, images and video and returns vectors. The catalogue files it under embedding.

Every current figure

Maker
Google
Register
model
Takes
text + image + video
Returns
embedding
Licence
not read

Known as 1 name

multimodalembedding