Struct HuggingFaceClient 

pub struct HuggingFaceClient { /* private fields */ }

Client for HuggingFace model hub and inference API

API Details

• Base URL: https://huggingface.co/api
• Rate limit: 30 requests/minute (free tier)
• API key optional for public models

Environment Variables

• HUGGINGFACE_API_KEY: Optional API key for higher rate limits and private models

Implementations

impl HuggingFaceClient

pub fn new() -> Self

Create a new HuggingFace client

Reads API key from HUGGINGFACE_API_KEY environment variable if available. Falls back to mock data if no API key is provided.

pub fn with_embedding_dim(embedding_dim: usize) -> Self

Create a new HuggingFace client with custom embedding dimension

pub async fn search_models( &self, query: &str, task: Option<&str>, ) -> Result<Vec<HuggingFaceModel>>

Search models by query and optional task filter

Arguments

• query - Search query string
• task - Optional task filter (e.g., “text-classification”, “text-generation”)

Example

let models = client.search_models("bert", Some("fill-mask")).await?;

pub async fn get_model( &self, model_id: &str, ) -> Result<Option<HuggingFaceModel>>

Get detailed information about a specific model

Arguments

• model_id - Model identifier (e.g., “bert-base-uncased”)

pub async fn list_datasets( &self, query: Option<&str>, ) -> Result<Vec<HuggingFaceDataset>>

List datasets with optional query filter

Arguments

• query - Optional search query for datasets

pub async fn get_dataset( &self, dataset_id: &str, ) -> Result<Option<HuggingFaceDataset>>

Get detailed information about a specific dataset

Arguments

• dataset_id - Dataset identifier

pub async fn inference( &self, model_id: &str, inputs: Value, ) -> Result<HuggingFaceInferenceResponse>

Run inference on a model

Arguments

• model_id - Model identifier
• inputs - Input data as JSON value

Note

Requires API key. Returns mock embeddings if no API key is available.

pub fn model_to_vector(&self, model: &HuggingFaceModel) -> SemanticVector

Convert HuggingFace model to SemanticVector

pub fn dataset_to_vector(&self, dataset: &HuggingFaceDataset) -> SemanticVector

Convert HuggingFace dataset to SemanticVector

Trait Implementations

impl Default for HuggingFaceClient

fn default() -> Self

Returns the “default value” for a type.