Physical benchmark sites
Each partner maintains a robot, workspace, task objects, cameras, and an evaluation protocol.
A PHYSICAL BENCHMARK NETWORK FOR ROBOT MANIPULATION
Your policy.
Our robot.
One comparable result.
RobotReplica is a network of organizations that host standardized robot manipulation benchmarks. Find a site with the same robot you use, send us your policy, and we evaluate it on maintained real-world tasks.
OVERVIEW
A shared evaluation service
for real robots.
Robotics results are difficult to compare when every lab uses a different robot, scene, and protocol. RobotReplica keeps physical benchmark sites running so the community can evaluate on consistent hardware and tasks.
Robot-matched evaluation
You choose the site with the same robot as your system. The site runs your policy for you.
Verified leaderboards
Every robot and site has its own board, keeping scores comparable and evidence traceable.
THE ROBOTREPLICA MODEL
Researchers do not need to rebuild the full benchmark. The benchmark stays at the host site; policies travel to it.
CURRENT SITES
Start with the robot
you already use.
Two organizations are building the first RobotReplica sites. Each site owns its hardware, tasks, evaluation process, and robot-specific leaderboard.
● Accepting evaluationsRichardson, Texas
IRVL @ UT Dallas
SO-101
RobotReplica SO-101 site
A hosted SO-101 evaluation site operated by IRVL at The University of Texas at Dallas.
- SO-101 arm
- UT Dallas
- Accepting evaluations
SITE 02● In developmentSan Francisco, California
General Intelligence Labs
OpenArm
RobotReplica OpenArm
A new hosted benchmark for the open-source OpenArm platform, extending the network to larger, bimanual, contact-rich manipulation.
- Bimanual platform
- Open-source hardware
- Protocol in development
HOW IT WORKS
From your model
to a verified score.
Find your robot
Choose a site that operates the same robot embodiment as your model or policy.
Contact the site
Share your policy, interface requirements, and the benchmark track you want to enter.
We run the evaluation
The host executes your policy on its maintained setup under a standardized protocol.
Compare the result
Your verified score is added to the leaderboard for that site and robot.
Evaluation details—policy interface, checkpoints, task coverage, and reporting—are coordinated directly with the selected host site.
LEADERBOARDS
Results grouped by
robot and site.
A score is meaningful only when the embodiment and physical protocol match. RobotReplica keeps those contexts explicit.
SO-101VLA-ReplicaIRVL @ UT DallasLIVEView leaderboard
OpenArmRobotReplica OpenArmGeneral Intelligence LabsIN DEVELOPMENTComing soon
EXPAND THE NETWORK
Operate a robot?
Host a site.
We are looking for organizations that can maintain a reproducible robot setup and evaluate community policies. Help bring a new embodiment into RobotReplica.
Propose a site