Dataset · Hugging Face

nvidia/OpenScienceReasoning-2

Name: OpenScienceReasoning-2
Creator: nvidia
License: cc-by-4.0
Keywords: license:cc-by-4.0, size_categories:100K<n<1M, format:parquet, modality:text, library:datasets, library:pandas, library:mlcroissant, library:polars, region:us

Dataset Description OpenScienceReasoning-2 is a multi-domain synthetic dataset designed to improve general-purpose reasoning in large language models (LLMs). The dataset contains multiple-choice and open-ended question-answer pairs with detailed reasoning traces and spans across diverse scientific domains, including STEM, law, economics, and humanities. OpenScience aims to boost accuracy on advanced benchmarks such as GPQA-Diamond, MMLU-Pro and HLE via supervised finetuning or… See the full description on the dataset page: https://huggingface.co/datasets/nvidia/OpenScienceReasoning-2.

View on Hugging FaceReport an issue

Author / organization	nvidia
License (as declared)	cc-by-4.0
Downloads	17.1K (as reported by Hugging Face)
Likes	56
Created	28 July 2025
Last updated at source	31 July 2025

Embed this entry

<iframe src="https://silytics.si/embed/item/6deba755-cc70-4402-a967-bfc86beabb9d" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

Free, no tracking. More about feeds and embeds.

More from nvidia

nvidia/PhysicalAI-Robotics-GR00T-X-Embodiment-SimDataset
nvidia/PhysicalAI-Autonomous-VehiclesDataset
nvidia/PhysicalAI-Robotics-Open-H-EmbodimentDataset
nvidia/SAGE-10kDataset
nvidia/simready-assetsDataset
nvidia/PhysicalAI-SmartSpacesDataset

Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 14 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.