Module embeddings 

Provider-agnostic embedding abstractions.

Embeddings are numerical representations of text or other inputs. Rig uses EmbeddingModel to generate vectors, Embed to select which text from a Rust value should be embedded, and EmbeddingsBuilder to batch embedding requests for vector stores or retrieval workflows.

Modules

builder

The module defines the EmbeddingsBuilder struct which accumulates objects to be embedded and batch generates the embeddings for each object when built. Only types that implement the Embed trait can be added to the EmbeddingsBuilder.

distance

Distance and similarity helpers for embedding vectors.

embed

The module defines the Embed trait, which must be implemented for types that can be embedded by the crate::embeddings::EmbeddingsBuilder.

embedding

The module defines the EmbeddingModel trait, which represents an embedding model that can generate embeddings for documents.

tool

The module defines the ToolSchema struct, which is used to embed an object that implements crate::tool::PortableToolEmbedding

Structs

EmbedError

Error type used for when the Embed::embed method of the Embed trait fails. Used by default implementations of Embed for common types.

Embedding

Struct that holds a single document and its embedding.

EmbeddingResponse

Response from an embedding request containing the embeddings and token usage.

EmbeddingsBuilder

Builder for creating embeddings from one or more documents of type T. Note: T can be any type that implements the Embed trait.

TextEmbedder

Accumulates string values that need to be embedded. Used by the Embed trait.

ToolSchema

Embeddable document that is used as an intermediate representation of a tool when RAGging tools.

Enums

EmbeddingError

Errors returned by embedding models.

Traits

Embed

Derive this trait for objects that need to be converted to vector embeddings. The Embed::embed method accumulates string values that need to be embedded by adding them to the TextEmbedder. If an error occurs, the method should return EmbedError.

EmbeddingModel

Trait for embedding models that can generate embeddings for documents.

ImageEmbeddingModel

Trait for embedding models that can generate embeddings for images.

Functions

to_texts

Utility function that returns a vector of strings that need to be embedded for a given object that implements the Embed trait.