Hugging Face· PlatformResearch paper

Jupyter Agents: training LLMs to reason with notebooks

Hugging Face introduces Jupyter Agents, a project to train small LLMs to reason with notebooks for data science tasks. They fine-tune Qwen3-4B on Kaggle notebooks and evaluate on DABStep benchmark, achieving improved accuracy after scaffolding optimization.

From the source

accuracy jumped from 44.4% → 59.7% (easy split).

huggingface.co

Announced

10 Sep 2025 00:00 UTC

Detected

03 Aug 2026 21:12 UTC (+327d after)