Embedding Models

Homepage

Mixpeek's video understanding, semantic embedding model

Index content from a specified URL with optional metadata and processing settings.

Upload and index files directly with optional metadata and processing settings.

Search for similar content using a URL, with support for various input types, modalities, and optional filters.

Search for similar content using an uploaded file, with optional filters and pagination.

Search indexed content using text queries with customizable filters and pagination.

Generates embeddings for text, image, audio, or video data to create machine-readable representations.

Generates descriptions for video content based on a given prompt.

Process a video given its URL or file path into chunks for easier embedding.

Process an image given its URL or file path into chunks for easier embedding.

Name	Modality	Dimensions	URL
`openai-clip-vit-base-patch32`	Text, Image	512	OpenAI Clip
`vuse-generic-v1`	Text, Video	768	Mixpeek