Dear colleagues,
We are inviting the OR/MS community to contribute problems to a new, community-driven benchmark for evaluating large language models (LLMs) on optimization modeling tasks.
LLMs are beginning to lower the barrier to entry for optimization modeling, but realizing that potential requires rigorous benchmarks that reflect what actually makes a modeling task hard — and those benchmarks are best shaped by our community. Inspired by community efforts such as Humanity's Last Exam, we are assembling an open, living benchmark of genuinely challenging optimization problems, alongside a paper documenting the collection and the performance of state-of-the-art LLMs on it.
The Management Science Editor-in-Chief is supportive of the project, and a department editor has agreed to an expedited review process (as with any submission, final publication depends on the review team's assessment). Contributors of accepted problems are invited to join the paper as co-authors.
We welcome problems from any application domain in operations and management science — production planning, supply chain and logistics, routing, inventory, facility location, scheduling, revenue management, energy systems, healthcare operations, and more. A good fit is a problem that is relevant to real-world operations and that current frontier LLMs struggle to model correctly.
Contributions are welcome through
August 31, 2026. Submissions are reviewed on a rolling basis, with a response typically within two weeks. Questions are very welcome — feel free to reach out at
or.bench2026@gmail.com.
We'd be grateful if you would forward this to colleagues who may be interested.
With thanks,
The organizing team
Jim Dai (Cornell University)
Dick den Hertog (University of Amsterdam)
Dongdong Ge (Shanghai Jiao Tong University)
Connor Lawless (Stanford University)
Kuo Liang (Shanghai Jiao Tong University)
Jianghao Lin (Shanghai Jiao Tong University)
Zi Ling (University of Chicago)
Jinsong Liu (Cornell University)
Hanzhang Qin (National University of Singapore)
Chung-Piaw Teo (National University of Singapore)
Madeleine Udell (Stanford University)
Wolfram Wiesemann (Imperial College London)
Ruihao Zhu (Cornell University)
Posted on 2026-06-12 by Wolfram Wiesemann