Expand description
Multimodal embedding types and traits (image / audio / mixed-modality).
These types accompany the trait definitions for embedders that consume
non-text inputs. Each trait sits alongside
EmbeddingModel and returns an
EmbedResult so remote providers can attach per-call usage counts.
Structs§
- Multimodal
Input - A heterogeneous input to a
MultimodalEmbeddingModel.
Enums§
- Audio
Input - An audio input to a transcription or audio embedding model.
- Modality
- Modality labels reported by
MultimodalEmbeddingModel::supported_modalities. - Multimodal
Block - One block of input to a
MultimodalEmbeddingModel.
Traits§
- Audio
Embedding Model - A model that produces dense vector embeddings from audio.
- Image
Embedding Model - A model that produces dense vector embeddings from images.
- Multimodal
Embedding Model - A model that produces a single dense vector from heterogeneous content (text + image + audio together).