Integrate with Hugging Face. Send prompts to any chat model served by Inference Providers, create text embeddings, classify and summarize text, generate images from a prompt, search the Hugging Face Hub for models and read a single model's metadata. Any other Hub or Inference Providers endpoint can be called directly.
Changelog
1 version
v1.0.0
Latest
- Initial release.
Actions
8
Classify Text
Classify a piece of text with a text-classification model served by Hugging Face Inference Providers, for example sentiment, topic or toxicity. Returns the highest scoring label plus all scored labels.
Create Embeddings
Turn a piece of text into a numeric embedding vector using a feature-extraction model served by Hugging Face Inference Providers.
Find Models
Search the Hugging Face Hub for model repositories by keyword, author, task or tag. Returns up to 100 models per call.
Generate Image
Generate an image from a text prompt using a text-to-image model served by Hugging Face Inference Providers. The image is saved to file storage and its file ID is returned.
Get Model
Get the metadata of a single model repository on the Hugging Face Hub by its repository ID.
Send Prompt
Send a prompt or a full chat conversation to any model served by Hugging Face Inference Providers and return the generated answer.
Summarize Text
Condense a long piece of text into a short summary using a summarization model served by Hugging Face Inference Providers.
Make API Call
Perform an arbitrary authorized API call to Hugging Face. Choose the Hub API or the Inference Providers router, then provide a path (or full URL), method, headers, query parameters and body. The access token is added automatically.