Hugging Face Model Catalogue

Explore the metadata for over 200,000 artificial intelligence models available on the Hugging Face platform. This data is essential for understanding the landscape of pre-trained models, tracking market trends, and aiding researchers, data scientists, and enthusiasts in their analysis of machine learning adoption and popularity.

Name: The model's unique identifier on Hugging Face. Author: The individual or organisation associated with the model. Repo Link: The direct hyperlink to the model’s repository on Hugging Face. Image URL: The web address for the model's icon or image. Category: Defines the model's type, such as Text Generation or Automatic Speech Recognition. Stars: The metric showing the number of endorsements the model has received. Downloads: The total usage count for the model. Last Updated: A timestamp indicating when the model was last modified.

The data is provided in a CSV file format, specifically huggingfacemodel.csv, and totals approximately 43.58 MB. The structure features 8 columns and contains over 201,000 valid records. Usability is rated highly at 10.00. Updates are anticipated on a monthly basis.

Ideal for data visualization of model popularity and growth trends. Researchers can use it for exploratory data analysis concerning model authorship and category adoption. Machine learning enthusiasts can track the success metrics (downloads and stars) of various model types. The data is also useful for comparative analysis across different fields of artificial intelligence.

The scope covers models listed on the popular Hugging Face platform. The temporal range of the update timestamps spans nearly a decade, with records ranging from August 2014 up to November 2023.

CC0: Public Domain

Data Scientists: For conducting market analysis of AI models and identifying emerging trends. Researchers: For studying the historical growth and categorization of pre-trained models. AI Enthusiasts: For discovering popular, highly-rated models and tracking recent updates. Developers: For finding frequently downloaded and proven resources to integrate into projects.

Original Data Source: Hugging Face Model Catalogue