Proudly built by:

AI2 Science Questions Mercury


The AI2 Science Questions Mercury dataset consists of questions used in student assessments across elementary and middle school grade levels, provided under license by an AI2 research partner. Each question is 4-way multiple choice format and may or may not include a diagram element. With 5,129 total questions without diagrams and 1,811 total questions with diagrams, this set can be used on its own or combined with the other science question sets AI2 offers.

  • Elementary School Without Diagrams Mercury v1.1: 1,434 questions
  • Elementary School With Diagrams Mercury v1.1: 832 questions
  • Middle School Without Diagrams Mercury v1.1: 3,695 questions
  • Middle School With Diagrams Mercury v1.1: 979 questions


This dataset was provided for distribution solely by AI2, and is provided for non-commercial research purposes only. Accessing and using this data requires adherence to the terms in our End User License Agreement, which you can find included in the download.


To evaluate your models, we have also built Aristo MINI, a light-weight question answering system that can quickly evaluate science questions with an evaluation web server and provided baseline solvers. You can extend the provided solvers with your own implementation to try out new approaches and compare results.

Release Notes

  • v. 1.1 (July 2017): Removes 12 extraneous questions (one Middle School NDMC Train, three Middle School NDMC Dev, and eight Middle School NDMC Test)

  • v. 1 was first released in November of 2016.


If you have any other questions or feedback for us about this data, please contact us at ai2-data@allenai.org.