Hugging Face Releases ml-intern: Open-Source AI Agent for Automated LLM Post-Training

Hugging Face launched ml-intern, an open-source AI agent built on its smolagents framework that automates the entire post-training workflow for large language models. The tool autonomously performs literature review, dataset discovery, training script execution, and iterative evaluation, demonstrating significant improvements by pushing a Qwen3-1.7B model's GPQA benchmark score from 8.5% to 32% in under 10 hours.