Skip to main content

Hugging Face

Use the Hugging Face connector to perform various operations such as managing models, datasets, and workflows. This connector allows you to interact with Hugging Face APIs directly from Boltic. You will need a valid API Key to get started.

Requirements​

To use the Hugging Face connector, you need:

AI Model Operations

Use AI Model Operations to perform a variety of tasks such as text processing, image analysis, and feature extraction. These operations help enhance applications with AI-powered capabilities like text generation, classification, translation, and more.

Prerequisites​

To use these AI operations, you need access to an AI model provider that supports these tasks. Depending on the provider, you may need an API key or OAuth authentication to access the APIs.

Supported Authentication Methods​

  • API Key

Basic Operations​

Chat Completion​

  • Generate a Chat Response: Provide a prompt to generate a contextual response using an AI chat model.

Feature Extraction​

  • Extract Features from Text or Image: Analyze input text or image and extract meaningful features.

Fill Mask​

  • Predict Missing Words: Provide a sentence with a missing word, and the model predicts the most likely word to fill the gap.

Image Classification​

  • Classify an Image: Assign labels to an image based on its contents.

Image Segmentation​

  • Segment Objects in an Image: Identify and segment different objects within an image.

Image-Text to Text​

  • Generate Text from an Image and Text Prompt: Provide both an image and a text prompt to generate meaningful output.

Object Detection​

  • Detect Objects in an Image: Identify objects within an image and return their locations.

Question Answering​

  • Answer Questions Based on Context: Provide a question and context, and the model generates an answer.

Summarization​

  • Summarize Text: Generate a concise summary from a longer piece of text.

Text Classification​

  • Categorize Text: Assign predefined categories or sentiments to a text input.

Text Generation​

  • Generate Text from a Prompt: Produce coherent and relevant text based on a given prompt.

Parameters​

Each operation requires specific parameters based on the task. Below is a list of available operations:

Chat Completion Operations​

  • Chat Completion: Generate a Chat Response

Feature Extraction Operations​

  • Feature extract: Extract Features from Text or Image

Fill Mask Operations​

  • Fill Mask: Predict Missing Words

Image Operations​

  • image.classify: Classify an Image
  • image.segment: Segment Objects in an Image
  • image.text-to-text: Generate Text from an Image and Text Prompt

Text Processing Operations​

  • text.qa: Answer Questions Based on Context
  • text.summarize: Summarize Text
  • text.classify: Categorize Text
  • text.generate: Generate Text from a Prompt
  • text.translate: Translate Text

Further Reading​

For more information on how to use these AI-powered operations, refer to the Hugging Face API documentation.