Datasets, competitions and hosted notebooks for machine learning
Kaggle is a community platform built around datasets and machine-learning competitions, owned by Google. Practitioners publish and download datasets, run notebooks in a hosted environment with free GPU or TPU sessions, and enter competitions scored on a public leaderboard. It also hosts a model hub for sharing checkpoints and short courses that walk through Python, machine learning and specific techniques. The differentiator is the surrounding community: discussion threads and public notebooks show how other people approached the same problem.
Last updated: 2026-09-20. This site only provides an index; for exact features, pricing, and licensing, see the official website.
dataset discovery, machine learning practice, competition entry and shared notebook experiments
If you're comparing similar products, check the alternatives below, or browse all tools in the AI Models & Platforms category.
No, but it helps. Hosted notebooks let beginners run other people's code immediately, and the short courses cover Python and machine learning basics. Competitions are harder: expect to read the discussion forum before writing any code.
Sessions provide free GPU or TPU time with limits on how long a run may last and how much time you consume each week. It is enough for coursework and small experiments, not for training large models repeatedly.
Sometimes, but licences vary per dataset and are set by the uploader. Read the licence field on the dataset page and check the original source before using anything in a product, because Kaggle hosting does not imply permission.
Developer console and API keys for the Claude model family
Node-based local interface for running image and video diffusion models
Browser playground for prompting and prototyping with Gemini models
Hosted Jupyter notebooks with optional free GPU and TPU runtimes