Hugging Face launches ML Intern in HuggingChat: an agent that runs your model training end-to-end

Thom_Wolf · x · 2026-09-10

Hugging Face co-founder Thom Wolf announced ML Intern in HuggingChat, aiming to make training a model for a task as easy as vibe-coding an app by 2026. The pitch: papers, datasets, pretrained models, benchmarks, storage and compute were already on the Hub — what was missing was an agent to assemble and run the whole process. You just ask; it handles the work, keeps you posted, and leaves you in control of steps and cost, finishing with a full set of shareable artifacts on the Hub.

Related event: Hugging Face Launches ML Intern Conversational Agent in HuggingChat(3 posts)→

Original post →

More from coding & agent

coding & agent channel →