new·The score now tells you which way it movedA brain's exam only ever grows: its own material writes questions, and so does every question a real caller asked and did not get answered. The score is a percentage over that growing set, so a brain that learned more could post a smaller number — and this week three did. One of them answered two MORE questions than the week before and showed eighteen points less. Printed as a single percentage, that reads as decline to a reader and as punishment to anyone who contributes material.all news →
mozg.beta
Sign in

AI SDK · Providers · all subjects

voyage-ai/capabilities

4 notes, read out of this brain and free to use. Each one was extracted from a source and is re-checked against its exam.

Voyage AI image embedding - multiple images

Multiple images can be embedded as a single embedding by passing an object with an `image` property containing an array of image URLs and/or base64 data URLs to embedMany().

Voyage AI multimodal embedding input format

Multimodal embeddings accept objects with `text` (array of strings) and `image` (array of image URLs or base64 data URLs) properties. Both text and image arrays can contain multiple values.

Voyage AI multimodal embedding constraints

The values list in multimodal embeddings must not contain more than 1,000 values. Each image must not exceed 16 million pixels or 20 MB in size. Each input must not exceed 32,000 tokens (with every 560 pixels of an image counted as a token). The total number of tokens across all inputs must not exceed 320,000.

Voyage AI supported image formats

Supported media types for image embeddings are: image/png, image/jpeg, image/webp, and image/gif. Images should be passed as data URL format: data:[mediatype];base64,<data>

Give your agent this brain