Enum EmbeddingProvider 

pub enum EmbeddingProvider {
    Onnx,
    Fastembed,
    OpenAi,
    OpenAiCompatible,
    Ollama,
    Mock,
}

Selects the backend used to generate embeddings.

Matches the Python SDK EMBEDDING_PROVIDER env var values so that a shared .env file works without modification across both SDKs.

Variants

Onnx

Local ONNX inference (existing engine). Equivalent to Python’s “fastembed” provider.

Fastembed

Alias for Onnx — accepts Python env files using EMBEDDING_PROVIDER=fastembed unchanged. Deserialized as Onnx internally.

OpenAi

OpenAI API or any OpenAI-compatible server (llama.cpp, vLLM, TEI, etc.). In Python, EMBEDDING_PROVIDER=openai uses LiteLLMEmbeddingEngine; in Rust we use a direct-HTTP engine. Both “openai” and “openai_compatible” map to the same engine.

OpenAiCompatible

Explicit alias for self-hosted OpenAI-compatible servers. Identical engine to OpenAi; exists so Python “openai_compatible” env files work unchanged.

Ollama

Ollama /api/embed endpoint (not OpenAI-compatible format).

Mock

Zero vectors; for testing. Activated by EMBEDDING_PROVIDER=mock or MOCK_EMBEDDING=true.

Trait Implementations

impl Clone for EmbeddingProvider

fn clone(&self) -> EmbeddingProvider

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for EmbeddingProvider

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Default for EmbeddingProvider

fn default() -> EmbeddingProvider

Returns the “default value” for a type.

impl<'de> Deserialize<'de> for EmbeddingProvider

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Eq for EmbeddingProvider

impl PartialEq for EmbeddingProvider

fn eq(&self, other: &EmbeddingProvider) -> bool

Tests for self and other values to be equal, and is used by ==.

fn ne(&self, other: &Rhs) -> bool

Tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl Serialize for EmbeddingProvider

fn serialize<__S>(&self, __serializer: __S) -> Result<__S::Ok, __S::Error>

where __S: Serializer,

Serialize this value into the given Serde serializer.

impl StructuralPartialEq for EmbeddingProvider