When “Done” Isn’t Done
Five models attempted a physical clearing and wiping task. Across 150 scored trials, none completed it. The partial results and completion claims tell a more revealing story.
Read articleThe Coop journal
Research & Insights
Notes on building, testing, and learning
from robots in the physical world.
Spur: A High-Level Controller for the Unitree R1
A high-level motor controller for the Unitree R1 did not exist. So we built it. Spur keeps the robot balanced while its arms and hands work, plans every motion before anything moves, and is now open source.
Read article
Coop’s First Benchmark: A Sign Wasn’t Enough
Four models, one humanoid, 220 trials. What a DO NOT ENTER sign revealed about instruction following, physical execution and reported success.
Read article
The 99% Problem: Why Robot Benchmarks Stopped Telling Us Anything
Why physical evaluation matters, where simulation falls short, and what benchmark saturation hides.
Read article
Why we’re building Coop
Independent, industry-specific benchmarks to help robot model developers understand what works, where it fails, and whether the next version is better.
Read article