# huggingface_client v0.1.0 - API Reference

## Modules

- [DepthEstimation](DepthEstimation.md): Estimate depth from an image (monocular depth estimation).
- [HuggingfaceClient](HuggingfaceClient.md): Top-level public API for the HuggingFace Elixir client.
- [HuggingfaceClient.Application](HuggingfaceClient.Application.md): OTP Application supervisor for HuggingfaceClient.
- [HuggingfaceClient.Client](HuggingfaceClient.Client.md): Represents a configured HuggingFace Inference client.
- [HuggingfaceClient.HardcodedMappings](HuggingfaceClient.HardcodedMappings.md): Agent-backed store for runtime dev overrides of provider/model mappings.
- [HuggingfaceClient.Hub.Compute.Jobs](HuggingfaceClient.Hub.Compute.Jobs.md): HuggingFace Hub Jobs API — run Docker workloads on HF infrastructure.
- [HuggingfaceClient.Hub.Compute.Training](HuggingfaceClient.Hub.Compute.Training.md): HuggingFace Training Stack — configuration helpers for fine-tuning and training.
- [HuggingfaceClient.Hub.Enterprise.Oauth](HuggingfaceClient.Hub.Enterprise.Oauth.md): OAuth PKCE flow for "Sign in with HuggingFace".
- [HuggingfaceClient.Hub.Enterprise.Security](HuggingfaceClient.Hub.Enterprise.Security.md): HuggingFace Enterprise Security & Compliance (7.2).
- [HuggingfaceClient.Hub.Hub](HuggingfaceClient.Hub.Hub.md): HuggingFace Hub — complete API client for the HF platform.
- [HuggingfaceClient.Hub.InferenceEndpoints](HuggingfaceClient.Hub.InferenceEndpoints.md): Manage Dedicated Inference Endpoints on the HuggingFace Hub.
- [HuggingfaceClient.Hub.Kernels](HuggingfaceClient.Hub.Kernels.md): HuggingFace Kernels API — load and run custom compute kernels from the Hub.
- [HuggingfaceClient.Hub.Metadata](HuggingfaceClient.Hub.Metadata.md): Model Cards and metadata management for HuggingFace repositories.
- [HuggingfaceClient.Hub.Mixins](HuggingfaceClient.Hub.Mixins.md): HuggingFace Hub Integration Mixins.
- [HuggingfaceClient.Hub.Models.Models](HuggingfaceClient.Hub.Models.Models.md): Model, dataset, and Space listing and info.
- [HuggingfaceClient.Hub.TensorBoard](HuggingfaceClient.Hub.TensorBoard.md): HuggingFace TensorBoard Logger — push TensorBoard logs to the Hub.
- [HuggingfaceClient.Hub.Users](HuggingfaceClient.Hub.Users.md): Community user profile APIs for the HuggingFace Hub.
- [HuggingfaceClient.Inference.Config](HuggingfaceClient.Inference.Config.md): Runtime configuration for the inference layer.
- [HuggingfaceClient.Inference.HTTPClient](HuggingfaceClient.Inference.HTTPClient.md): Behaviour for the underlying HTTP client, enabling Mox-based testing.
- [HuggingfaceClient.Inference.HttpHelper](HuggingfaceClient.Inference.HttpHelper.md): Shared HTTP POST helper for TGI and TEI dedicated server clients.
- [HuggingfaceClient.Inference.Inference](HuggingfaceClient.Inference.Inference.md): Elixir client for the Hugging Face Inference API.
- [HuggingfaceClient.Inference.ModelInfo](HuggingfaceClient.Inference.ModelInfo.md): Fetches model metadata and provider mapping information from the HuggingFace Hub.
- [HuggingfaceClient.Inference.SSE](HuggingfaceClient.Inference.SSE.md): Server-Sent Events (SSE) parser for streaming inference responses.
- [HuggingfaceClient.Inference.StreamHelpers](HuggingfaceClient.Inference.StreamHelpers.md): Utility functions for working with streaming inference responses.
- [HuggingfaceClient.Inference.TEI](HuggingfaceClient.Inference.TEI.md): Client for HuggingFace Text Embeddings Inference (TEI) servers.
- [HuggingfaceClient.Inference.Telemetry](HuggingfaceClient.Inference.Telemetry.md): Telemetry integration for `HuggingfaceClient`.
- [HuggingfaceClient.Inference.Types](HuggingfaceClient.Inference.Types.md): Shared type definitions for the HuggingFace Inference API.
- [HuggingfaceClient.Jinja.Jinja](HuggingfaceClient.Jinja.Jinja.md): Jinja2 template engine for HuggingFace chat templates.
- [HuggingfaceClient.Libraries](HuggingfaceClient.Libraries.md): HuggingFace Library Integration Helpers.
- [HuggingfaceClient.MCP](HuggingfaceClient.MCP.md): HuggingFace MCP Client — Model Context Protocol integration.
- [HuggingfaceClient.Plug](HuggingfaceClient.Plug.md): Optional Plug middleware for Phoenix / Plug applications.
- [HuggingfaceClient.Provider](HuggingfaceClient.Provider.md): Behaviour that every inference provider module must implement.
- [HuggingfaceClient.Provider.Baseten](HuggingfaceClient.Provider.Baseten.md): Baseten – client-side routing, conversational task.
- [HuggingfaceClient.Provider.BlackForestLabs](HuggingfaceClient.Provider.BlackForestLabs.md): Black Forest Labs – text-to-image task (client-side routing).
- [HuggingfaceClient.Provider.Cerebras](HuggingfaceClient.Provider.Cerebras.md): Cerebras – conversational task.
- [HuggingfaceClient.Provider.Clarifai](HuggingfaceClient.Provider.Clarifai.md): Clarifai – conversational task.
- [HuggingfaceClient.Provider.Cohere](HuggingfaceClient.Provider.Cohere.md): Cohere – conversational task.
- [HuggingfaceClient.Provider.DeepInfra](HuggingfaceClient.Provider.DeepInfra.md): DeepInfra – conversational task.
- [HuggingfaceClient.Provider.FalAi](HuggingfaceClient.Provider.FalAi.md): Fal.ai – multiple tasks including image/video/speech.
- [HuggingfaceClient.Provider.FeatherlessAi](HuggingfaceClient.Provider.FeatherlessAi.md): Featherless AI – conversational task.
- [HuggingfaceClient.Provider.FireworksAi](HuggingfaceClient.Provider.FireworksAi.md): Fireworks AI – conversational task.
- [HuggingfaceClient.Provider.Groq](HuggingfaceClient.Provider.Groq.md): Groq inference provider (conversational / chat completions).
- [HuggingfaceClient.Provider.GroqTextGeneration](HuggingfaceClient.Provider.GroqTextGeneration.md): Groq provider – text-generation task.
- [HuggingfaceClient.Provider.HfInference](HuggingfaceClient.Provider.HfInference.md): Provider implementation for HuggingFace's own inference router / API.
- [HuggingfaceClient.Provider.Hyperbolic](HuggingfaceClient.Provider.Hyperbolic.md): Hyperbolic – conversational task.
- [HuggingfaceClient.Provider.HyperbolicTextToImage](HuggingfaceClient.Provider.HyperbolicTextToImage.md): Hyperbolic – text-to-image task.
- [HuggingfaceClient.Provider.Nebius](HuggingfaceClient.Provider.Nebius.md): Nebius Studio – conversational task.
- [HuggingfaceClient.Provider.NebiusFeatureExtraction](HuggingfaceClient.Provider.NebiusFeatureExtraction.md): Nebius Studio – feature-extraction / embeddings task.
- [HuggingfaceClient.Provider.NebiusTextToImage](HuggingfaceClient.Provider.NebiusTextToImage.md): Nebius Studio – text-to-image task.
- [HuggingfaceClient.Provider.Novita](HuggingfaceClient.Provider.Novita.md): Novita AI – conversational task.
- [HuggingfaceClient.Provider.NovitaTextToVideo](HuggingfaceClient.Provider.NovitaTextToVideo.md): Novita AI – text-to-video task.
- [HuggingfaceClient.Provider.Nscale](HuggingfaceClient.Provider.Nscale.md): Nscale – conversational task.
- [HuggingfaceClient.Provider.NscaleTextToImage](HuggingfaceClient.Provider.NscaleTextToImage.md): Nscale – text-to-image task.
- [HuggingfaceClient.Provider.Nvidia](HuggingfaceClient.Provider.Nvidia.md): NVIDIA NIM – conversational task.
- [HuggingfaceClient.Provider.OpenAi](HuggingfaceClient.Provider.OpenAi.md): OpenAI – conversational task.
- [HuggingfaceClient.Provider.OvhCloud](HuggingfaceClient.Provider.OvhCloud.md): OVHcloud – conversational task.
- [HuggingfaceClient.Provider.PublicAi](HuggingfaceClient.Provider.PublicAi.md): PublicAI – conversational task.
- [HuggingfaceClient.Provider.Replicate](HuggingfaceClient.Provider.Replicate.md): Replicate – client-side routing, multiple tasks.
- [HuggingfaceClient.Provider.Sambanova](HuggingfaceClient.Provider.Sambanova.md): SambaNova – conversational task.
- [HuggingfaceClient.Provider.SambanovaFeatureExtraction](HuggingfaceClient.Provider.SambanovaFeatureExtraction.md): SambaNova – feature-extraction task.
- [HuggingfaceClient.Provider.Scaleway](HuggingfaceClient.Provider.Scaleway.md): Scaleway – conversational task.
- [HuggingfaceClient.Provider.Together](HuggingfaceClient.Provider.Together.md): Together AI – conversational task.
- [HuggingfaceClient.Provider.TogetherTextGeneration](HuggingfaceClient.Provider.TogetherTextGeneration.md): Together AI – text-generation task.
- [HuggingfaceClient.Provider.TogetherTextToImage](HuggingfaceClient.Provider.TogetherTextToImage.md): Together AI – text-to-image task.
- [HuggingfaceClient.Provider.Wavespeed](HuggingfaceClient.Provider.Wavespeed.md): Wavespeed – text-to-image and text-to-video (client-side routing).
- [HuggingfaceClient.Provider.ZaiOrg](HuggingfaceClient.Provider.ZaiOrg.md): Zai.org – conversational task.
- [HuggingfaceClient.ProviderRegistry](HuggingfaceClient.ProviderRegistry.md): Lookup table mapping `{provider_id, task}` pairs to provider module atoms.
- [HuggingfaceClient.Request](HuggingfaceClient.Request.md): Core HTTP execution layer for all inference requests.
- [HuggingfaceClient.ResponseHandler](HuggingfaceClient.ResponseHandler.md): Resolves provider response tuples, including async/polling patterns used by
providers like Fal.ai, Novita, and Black Forest Labs.
- [HuggingfaceClient.SSE](HuggingfaceClient.SSE.md): Server-Sent Events (SSE) stream parser following the W3C EventSource specification.
- [HuggingfaceClient.Serialization](HuggingfaceClient.Serialization.md): HuggingFace Serialization — save and load ML model weights.
- [HuggingfaceClient.Serialization.DDUF](HuggingfaceClient.Serialization.DDUF.md): DDUF (Diffusion model Distributed Unified Format) file operations.
- [HuggingfaceClient.StreamHelpers](HuggingfaceClient.StreamHelpers.md): Utilities for consuming streaming chat-completion responses.
- [ImageTextToText](ImageTextToText.md): Multimodal vision-language models (VLMs).
- [MaskGeneration](MaskGeneration.md): Generate segmentation masks (SAM-style). Returns masks for objects in an image.
- [VideoClassification](VideoClassification.md): Classify a video clip into predefined categories.

- Inference — Providers
  - [HuggingfaceClient.Inference.Provider](HuggingfaceClient.Inference.Provider.md): Behaviour that every inference provider module must implement.
  - [HuggingfaceClient.Inference.Provider.BaseConversational](HuggingfaceClient.Inference.Provider.BaseConversational.md): Base implementation for providers that use the OpenAI-compatible chat completions endpoint.
  - [HuggingfaceClient.Inference.Provider.BaseTextGeneration](HuggingfaceClient.Inference.Provider.BaseTextGeneration.md): Base implementation for providers that use an OpenAI-compatible completions endpoint.
  - [HuggingfaceClient.Inference.Provider.Baseten](HuggingfaceClient.Inference.Provider.Baseten.md): Baseten inference provider.
  - [HuggingfaceClient.Inference.Provider.BlackForestLabs](HuggingfaceClient.Inference.Provider.BlackForestLabs.md): Black Forest Labs (FLUX) inference provider.
  - [HuggingfaceClient.Inference.Provider.Cerebras](HuggingfaceClient.Inference.Provider.Cerebras.md): Cerebras inference provider.
  - [HuggingfaceClient.Inference.Provider.Clarifai](HuggingfaceClient.Inference.Provider.Clarifai.md): Clarifai inference provider.
  - [HuggingfaceClient.Inference.Provider.Cohere](HuggingfaceClient.Inference.Provider.Cohere.md): Cohere inference provider.
  - [HuggingfaceClient.Inference.Provider.DeepInfra](HuggingfaceClient.Inference.Provider.DeepInfra.md): DeepInfra inference provider.
  - [HuggingfaceClient.Inference.Provider.FalAi](HuggingfaceClient.Inference.Provider.FalAi.md): Fal.ai inference provider. Uses async queue polling for image/video tasks.
  - [HuggingfaceClient.Inference.Provider.FeatherlessAi](HuggingfaceClient.Inference.Provider.FeatherlessAi.md): Featherless.ai inference provider.
  - [HuggingfaceClient.Inference.Provider.FireworksAi](HuggingfaceClient.Inference.Provider.FireworksAi.md): Fireworks AI inference provider.
  - [HuggingfaceClient.Inference.Provider.Groq](HuggingfaceClient.Inference.Provider.Groq.md): Groq inference provider. Supports conversational and text-generation tasks.
  - [HuggingfaceClient.Inference.Provider.HfInference](HuggingfaceClient.Inference.Provider.HfInference.md): Provider implementation for HuggingFace's own serverless Inference API.
  - [HuggingfaceClient.Inference.Provider.Hyperbolic](HuggingfaceClient.Inference.Provider.Hyperbolic.md): Hyperbolic inference provider.
  - [HuggingfaceClient.Inference.Provider.Nebius](HuggingfaceClient.Inference.Provider.Nebius.md): Nebius AI inference provider.
  - [HuggingfaceClient.Inference.Provider.Novita](HuggingfaceClient.Inference.Provider.Novita.md): Novita AI inference provider.
  - [HuggingfaceClient.Inference.Provider.Nscale](HuggingfaceClient.Inference.Provider.Nscale.md): Nscale inference provider.
  - [HuggingfaceClient.Inference.Provider.Nvidia](HuggingfaceClient.Inference.Provider.Nvidia.md): NVIDIA inference provider.
  - [HuggingfaceClient.Inference.Provider.OpenAi](HuggingfaceClient.Inference.Provider.OpenAi.md): OpenAI inference provider.
  - [HuggingfaceClient.Inference.Provider.OvhCloud](HuggingfaceClient.Inference.Provider.OvhCloud.md): OVHcloud inference provider.
  - [HuggingfaceClient.Inference.Provider.PublicAi](HuggingfaceClient.Inference.Provider.PublicAi.md): PublicAI inference provider.
  - [HuggingfaceClient.Inference.Provider.Replicate](HuggingfaceClient.Inference.Provider.Replicate.md): Replicate inference provider.
  - [HuggingfaceClient.Inference.Provider.Sambanova](HuggingfaceClient.Inference.Provider.Sambanova.md): SambaNova inference provider.
  - [HuggingfaceClient.Inference.Provider.Scaleway](HuggingfaceClient.Inference.Provider.Scaleway.md): Scaleway inference provider.
  - [HuggingfaceClient.Inference.Provider.Together](HuggingfaceClient.Inference.Provider.Together.md): Together.ai inference provider.
  - [HuggingfaceClient.Inference.Provider.Wavespeed](HuggingfaceClient.Inference.Provider.Wavespeed.md): Wavespeed AI inference provider.
  - [HuggingfaceClient.Inference.Provider.ZaiOrg](HuggingfaceClient.Inference.Provider.ZaiOrg.md): Zai.com inference provider.
  - [HuggingfaceClient.Inference.ProviderMapping](HuggingfaceClient.Inference.ProviderMapping.md): ETS-backed GenServer that caches `inferenceProviderMapping` data fetched
from the HuggingFace Hub API (`GET /api/models/:model_id`).

- Inference — Tasks
  - [HuggingfaceClient.Inference.Task](HuggingfaceClient.Inference.Task.md): Shared helpers for task modules.
  - [HuggingfaceClient.Inference.Task.AudioClassification](HuggingfaceClient.Inference.Task.AudioClassification.md): Audio classification. Returns label + score pairs.
  - [HuggingfaceClient.Inference.Task.AudioToAudio](HuggingfaceClient.Inference.Task.AudioToAudio.md): Audio-to-audio transformation (source separation, enhancement).
  - [HuggingfaceClient.Inference.Task.AutomaticSpeechRecognition](HuggingfaceClient.Inference.Task.AutomaticSpeechRecognition.md): ASR — transcribes audio to text.
  - [HuggingfaceClient.Inference.Task.ChatCompletion](HuggingfaceClient.Inference.Task.ChatCompletion.md): Chat completion task — OpenAI-compatible `/v1/chat/completions`.
  - [HuggingfaceClient.Inference.Task.DocumentQuestionAnswering](HuggingfaceClient.Inference.Task.DocumentQuestionAnswering.md): Document question answering from scanned documents.
  - [HuggingfaceClient.Inference.Task.FeatureExtraction](HuggingfaceClient.Inference.Task.FeatureExtraction.md): Dense embedding / feature extraction (returns float arrays).
  - [HuggingfaceClient.Inference.Task.FillMask](HuggingfaceClient.Inference.Task.FillMask.md): Fill-mask (masked language modeling) task.
  - [HuggingfaceClient.Inference.Task.ImageClassification](HuggingfaceClient.Inference.Task.ImageClassification.md): Image classification. Returns `[%{"label" => ..., "score" => ...}]`.
  - [HuggingfaceClient.Inference.Task.ImageSegmentation](HuggingfaceClient.Inference.Task.ImageSegmentation.md): Image segmentation. Returns masks with labels and scores.
  - [HuggingfaceClient.Inference.Task.ImageTextToImage](HuggingfaceClient.Inference.Task.ImageTextToImage.md): Image + text to image. Takes an image and a text prompt, returns a new image.
  - [HuggingfaceClient.Inference.Task.ImageTextToVideo](HuggingfaceClient.Inference.Task.ImageTextToVideo.md): Image + text to video. Takes an image and a text prompt, returns a video.
  - [HuggingfaceClient.Inference.Task.ImageToImage](HuggingfaceClient.Inference.Task.ImageToImage.md): Image-to-image transformation (e.g. style transfer, super-resolution).
  - [HuggingfaceClient.Inference.Task.ImageToText](HuggingfaceClient.Inference.Task.ImageToText.md): Image captioning / visual to text.
  - [HuggingfaceClient.Inference.Task.ImageToVideo](HuggingfaceClient.Inference.Task.ImageToVideo.md): Animates a still image into a short video clip.
  - [HuggingfaceClient.Inference.Task.ObjectDetection](HuggingfaceClient.Inference.Task.ObjectDetection.md): Object detection with bounding boxes.
  - [HuggingfaceClient.Inference.Task.QuestionAnswering](HuggingfaceClient.Inference.Task.QuestionAnswering.md): Extractive question answering from context.
  - [HuggingfaceClient.Inference.Task.SentenceSimilarity](HuggingfaceClient.Inference.Task.SentenceSimilarity.md): Sentence similarity scoring (returns a list of similarity scores).
  - [HuggingfaceClient.Inference.Task.Summarization](HuggingfaceClient.Inference.Task.Summarization.md): Abstractive text summarization.
  - [HuggingfaceClient.Inference.Task.TableQuestionAnswering](HuggingfaceClient.Inference.Task.TableQuestionAnswering.md): Question answering over tabular data (TAPAS-style).
  - [HuggingfaceClient.Inference.Task.TabularClassification](HuggingfaceClient.Inference.Task.TabularClassification.md): Tabular data classification. Returns class indices.
  - [HuggingfaceClient.Inference.Task.TabularRegression](HuggingfaceClient.Inference.Task.TabularRegression.md): Tabular data regression. Returns predicted float values.
  - [HuggingfaceClient.Inference.Task.TextClassification](HuggingfaceClient.Inference.Task.TextClassification.md): Text classification task. Returns label + score pairs.
  - [HuggingfaceClient.Inference.Task.TextGeneration](HuggingfaceClient.Inference.Task.TextGeneration.md): Text generation (completion) task.
  - [HuggingfaceClient.Inference.Task.TextToAudio](HuggingfaceClient.Inference.Task.TextToAudio.md): Text-to-audio generation (music, sound effects). Returns audio bytes.
  - [HuggingfaceClient.Inference.Task.TextToImage](HuggingfaceClient.Inference.Task.TextToImage.md): Text-to-image generation.
  - [HuggingfaceClient.Inference.Task.TextToSpeech](HuggingfaceClient.Inference.Task.TextToSpeech.md): Text-to-speech synthesis. Returns audio bytes.
  - [HuggingfaceClient.Inference.Task.TextToVideo](HuggingfaceClient.Inference.Task.TextToVideo.md): Text-to-video generation. Returns video bytes.
  - [HuggingfaceClient.Inference.Task.TokenClassification](HuggingfaceClient.Inference.Task.TokenClassification.md): Named entity recognition / token classification.
  - [HuggingfaceClient.Inference.Task.Translation](HuggingfaceClient.Inference.Task.Translation.md): Neural machine translation.
  - [HuggingfaceClient.Inference.Task.VisualQuestionAnswering](HuggingfaceClient.Inference.Task.VisualQuestionAnswering.md): Visual question answering (image + question → answer).
  - [HuggingfaceClient.Inference.Task.ZeroShotClassification](HuggingfaceClient.Inference.Task.ZeroShotClassification.md): Zero-shot text classification with candidate labels.
  - [HuggingfaceClient.Inference.Task.ZeroShotImageClassification](HuggingfaceClient.Inference.Task.ZeroShotImageClassification.md): Zero-shot image classification with candidate labels.
  - [HuggingfaceClient.Inference.TaskRunner](HuggingfaceClient.Inference.TaskRunner.md): Shared run/stream/resolve_provider logic used by all inference task modules.

- Inference — TGI
  - [HuggingfaceClient.Inference.TGI](HuggingfaceClient.Inference.TGI.md): Client for HuggingFace Text Generation Inference (TGI) servers.

- Hub — Core
  - [HuggingfaceClient.Hub](HuggingfaceClient.Hub.md): Top-level facade for all HuggingFace Hub operations.
  - [HuggingfaceClient.Hub.Auth](HuggingfaceClient.Hub.Auth.md): Authentication and user info (`/api/whoami-v2`).
  - [HuggingfaceClient.Hub.Client](HuggingfaceClient.Hub.Client.md): Core HTTP client for HuggingFace Hub API interactions.

- Hub — Repositories
  - [HuggingfaceClient.Hub.Commits](HuggingfaceClient.Hub.Commits.md): High-level commit helpers: upload files, delete files.
  - [HuggingfaceClient.Hub.Files](HuggingfaceClient.Hub.Files.md): File listing, existence checks, downloads, paths-info, and safetensors parsing.
  - [HuggingfaceClient.Hub.Repos](HuggingfaceClient.Hub.Repos.md): Repository management on the HuggingFace Hub.
  - [HuggingfaceClient.Hub.Repositories.Repos](HuggingfaceClient.Hub.Repositories.Repos.md): Repository creation, deletion, branch management, and commit history.
  - [HuggingfaceClient.Hub.Repositories.Snapshots](HuggingfaceClient.Hub.Repositories.Snapshots.md): Snapshot download — mirrors `snapshotDownload` from `@huggingface/hub`.
  - [HuggingfaceClient.Hub.Snapshots](HuggingfaceClient.Hub.Snapshots.md): Cache path utilities for HuggingFace Hub snapshots.
  - [HuggingfaceClient.Hub.Tags](HuggingfaceClient.Hub.Tags.md): Manage git tags and list refs (branches/tags) on HuggingFace repositories.

- Hub — Models
  - [HuggingfaceClient.Hub.Models](HuggingfaceClient.Hub.Models.md): Search and retrieve model metadata from the HuggingFace Hub.

- Hub — Datasets
  - [HuggingfaceClient.Hub.DatasetViewer](HuggingfaceClient.Hub.DatasetViewer.md): HuggingFace Dataset Viewer API.
  - [HuggingfaceClient.Hub.Datasets](HuggingfaceClient.Hub.Datasets.md): HuggingFace Datasets Hub — access and manage dataset repositories.
Delegates to `HuggingfaceClient.Hub.Models` which handles all repo types.

- Hub — Spaces
  - [HuggingfaceClient.Hub.Spaces](HuggingfaceClient.Hub.Spaces.md): Manage Hugging Face Spaces — hosted ML applications.

- Hub — Discovery
  - [HuggingfaceClient.Hub.Collections](HuggingfaceClient.Hub.Collections.md): CRUD operations for HuggingFace Hub Collections.
  - [HuggingfaceClient.Hub.Discovery.Collections](HuggingfaceClient.Hub.Discovery.Collections.md): Collections API — list, create, get, add/remove items, delete.
  - [HuggingfaceClient.Hub.Papers](HuggingfaceClient.Hub.Papers.md): Access and search research papers on the HuggingFace Hub.
  - [HuggingfaceClient.Hub.Search](HuggingfaceClient.Hub.Search.md): Enhanced search across models, datasets, and spaces on the HuggingFace Hub.

- Hub — Collaboration
  - [HuggingfaceClient.Hub.Discussions](HuggingfaceClient.Hub.Discussions.md): Discussions and Pull Requests on the HuggingFace Hub.
  - [HuggingfaceClient.Hub.Webhooks](HuggingfaceClient.Hub.Webhooks.md): Manage webhooks on the HuggingFace Hub.
  - [HuggingfaceClient.Hub.WebhooksServer](HuggingfaceClient.Hub.WebhooksServer.md): HuggingFace Webhook payload handling and signature verification.

- Hub — Storage
  - [HuggingfaceClient.Hub.Buckets](HuggingfaceClient.Hub.Buckets.md): HuggingFace Storage Buckets — S3-like object storage powered by Xet.
  - [HuggingfaceClient.Hub.Cache](HuggingfaceClient.Hub.Cache.md): HuggingFace Hub local cache management.
  - [HuggingfaceClient.Hub.FileSystem](HuggingfaceClient.Hub.FileSystem.md): HuggingFace FileSystem — fsspec-compatible interface to Hub repos and Buckets.

- Hub — Compute
  - [HuggingfaceClient.Hub.AutoTrain](HuggingfaceClient.Hub.AutoTrain.md): HuggingFace AutoTrain API — no-code/low-code model training.
  - [HuggingfaceClient.Hub.Jobs](HuggingfaceClient.Hub.Jobs.md): HuggingFace Hub Inference Jobs API.

- Hub — Enterprise
  - [HuggingfaceClient.Hub.GatedRepos](HuggingfaceClient.Hub.GatedRepos.md): Manage access requests for gated repositories on the HuggingFace Hub.
  - [HuggingfaceClient.Hub.OAuth](HuggingfaceClient.Hub.OAuth.md): PKCE OAuth 2.0 login flow for HuggingFace Hub.
  - [HuggingfaceClient.Hub.Organizations](HuggingfaceClient.Hub.Organizations.md): HuggingFace Organization management.

- Hub — Evaluation
  - [HuggingfaceClient.Hub.Evaluate](HuggingfaceClient.Hub.Evaluate.md): HuggingFace Evaluate — metrics computation API.
  - [HuggingfaceClient.Hub.Leaderboards](HuggingfaceClient.Hub.Leaderboards.md): HuggingFace Hub Leaderboards API.

- Utilities
  - [HuggingfaceClient.Agent](HuggingfaceClient.Agent.md): Agentic chat loop with tool use — an Elixir port of `@huggingface/mcp-client` Agent.
  - [HuggingfaceClient.Config](HuggingfaceClient.Config.md): HuggingFace Client configuration and environment variable management.
  - [HuggingfaceClient.Error](HuggingfaceClient.Error.md): Base exception for all HuggingFace Inference errors.
  - [HuggingfaceClient.Jinja](HuggingfaceClient.Jinja.md): A Jinja2-compatible template engine for rendering HuggingFace chat templates.
  - [HuggingfaceClient.Telemetry](HuggingfaceClient.Telemetry.md): Telemetry integration for HuggingfaceClient inference requests.

- Exceptions
  - [HuggingfaceClient.Error.HttpRequestError](HuggingfaceClient.Error.HttpRequestError.md): Base for errors that carry HTTP request/response context.
The `Authorization` header is automatically redacted.

  - [HuggingfaceClient.Error.HubApiError](HuggingfaceClient.Error.HubApiError.md): Raised when the HTTP request to the Hugging Face Hub fails.
  - [HuggingfaceClient.Error.InputError](HuggingfaceClient.Error.InputError.md): Raised when the caller supplies invalid arguments.
  - [HuggingfaceClient.Error.ProviderApiError](HuggingfaceClient.Error.ProviderApiError.md): Raised when the HTTP request to the provider fails (4xx/5xx).
  - [HuggingfaceClient.Error.ProviderOutputError](HuggingfaceClient.Error.ProviderOutputError.md): Raised when the provider returns a response that doesn't match expectations.
  - [HuggingfaceClient.Error.RoutingError](HuggingfaceClient.Error.RoutingError.md): Raised when the inference request cannot be routed to any provider.

## Mix Tasks

- [mix hf.download](Mix.Tasks.Hf.Download.md): Downloads a file or an entire repository snapshot from the HuggingFace Hub.
- [mix hf.login](Mix.Tasks.Hf.Login.md): Saves a HuggingFace API token to `~/.cache/huggingface/token`.
- [mix hf.logout](Mix.Tasks.Hf.Logout.md): Removes the saved HuggingFace API token.
- [mix hf.model_info](Mix.Tasks.Hf.ModelInfo.md): Fetches metadata for a model from the HuggingFace Hub and displays it,
including all available inference providers.
- [mix hf.providers](Mix.Tasks.Hf.Providers.md): Lists all providers registered in `HuggingfaceClient.Inference.ProviderRegistry`.
- [mix hf.repos](Mix.Tasks.Hf.Repos.md): Lists repositories for the authenticated user or a specific author.
- [mix hf.spaces](Mix.Tasks.Hf.Spaces.md): List and manage HuggingFace Spaces.
- [mix hf.upload](Mix.Tasks.Hf.Upload.md): Uploads a file or folder to a HuggingFace Hub repository.
- [mix hf.whoami](Mix.Tasks.Hf.Whoami.md): Displays information about the currently authenticated user.

- Hub — Storage
  - [mix hf.cache](Mix.Tasks.Hf.Cache.md): Manage the local HuggingFace Hub cache directory.