Module embedding 

Embedding engine abstraction and providers.

Structs

EmbeddingConfig

Unified embedding configuration.

MockEmbeddingEngine

A mock embedding engine.

ModelUrls

HuggingFace Hub URLs for supported models

OllamaEmbeddingEngine

Embedding engine that calls the Ollama /api/embed HTTP endpoint.

OnnxEmbeddingConfig

ONNX-specific configuration.

OnnxEmbeddingEngine

ONNX-based embedding engine for local inference

OpenAICompatibleEmbeddingEngine

Embedding engine that calls an OpenAI-compatible /v1/embeddings HTTP endpoint.

Enums

EmbeddingError

Error type for embedding engine operations.

EmbeddingProvider

Selects the backend used to generate embeddings.

Traits

EmbeddingEngine

Core trait for text embedding engines

Functions

download_model

Download both model and tokenizer for a specific configuration.

ensure_model_exists

Ensure a model file exists, downloading it if necessary.

ensure_tokenizer_exists

Ensure a tokenizer file exists, downloading it if necessary.

handle_embedding_response

Replace embeddings for originally-invalid inputs with zero vectors.

is_embeddable

Returns true if s is non-empty after stripping ASCII whitespace.

sanitize_embedding_inputs

Replace empty/whitespace-only strings with "." to prevent API errors.

Type Aliases

EmbeddingResult

Convenience Result alias for embedding engine operations.