Hugging Face has launched ML Intern, a chatbot feature that allows users to execute machine learning experiments without prior technical expertise. A user describes their idea in conversation, prompting the assistant to search the Hugging Face Hub, GitHub, and the web for suitable models, datasets, and tools. Before execution begins, the system estimates required compute costs and proposes a budget which the user must approve. Once approved, the tool operates autonomously to create datasets, train models, monitor jobs, upload results to the Hub, write reports, and build demos. Each training run receives its own dashboard for tracking progress. One example from the demo video ran for approximately six hours and cost less than $0.50.
This tool lowers the barrier for new projects on the platform by handling complex workflows through simple dialogue. It addresses the gap between having an idea and possessing the skills to implement it. The launch occurs as Hugging Face prepares for an acquisition by Nvidia, with CEO Jensen Huang promising to keep the platform open and hardware-neutral.
* Costs stay within the user-defined budget limit
* One demo run lasted six hours under half a dollar
* Results upload directly to the public Hub
