Mixedbread Wholembed v3

Mixedbread Wholembed v3 is a unified omnimodal multilingual late-interaction retrieval model that sets a new state-of-the-art across text, audio, images, and video. It is the first semantic model to outperform lexical-based retrieval on LIMIT benchmark and powers Mixedbread Search as the default model.

Mixedbread Wholembed v3 is a unified omnimodal multilingual late-interaction retrieval model that sets a new state-of-the-art across text, audio, images, and video. It is the first semantic model to outperform lexical-based retrieval on LIMIT benchmark and powers Mixedbread Search as the default model.

• 01First semantic retrieval model to surpass lexical BM25 on the LIMIT benchmark
• 02Unified retrieval across text, audio, images, and video in 100+ languages
• 032M free tokens for new users; default model for all new Mixedbread Search stores