Robocurve
Evaluating robots in the real world.
About Robocurve
Robocurve builds open-source tools and independent benchmarks to measure how well robots can do real-world jobs. Instead of relying on unverified demo videos from frontier labs, we score their models on reproducible benchmarks that anyone can trust. Today there are no well-run, standardized robotics benchmarks. Labs evaluate in-house, and no independent group has stepped in to run continuous benchmarking as a service. The result is that no one actually knows how good anyone else is, or where the real frontier sits. Good benchmarks require operating and maintaining physical hardware and real-world setups. Simulation only goes so far, since models that look strong in sim can show large performance gaps once deployed in the real world. Building real-world benchmarks means coordinating job-domain experts, evals engineering, and hands-on robotics all at once. Frontier labs are targeting general-purpose robotics by 2028, yet the field of robotics evals barely exists. Whoever builds the trusted measure of robot capability becomes the reference everyone relies on. We combine backgrounds in AI evals and robotics to build and grow this field as fast as possible. We already shipped v1 of our open-source framework, Inspect Robots, and ran our first pilot scoring a frontier model on a real robot.
Public traction evidence
Each signal links to the public source used for attribution.
- X
Robots can bake bread, cut carrots, and do karate kicks in demo videos. But how capable are they, really? @robocurve measures how good A...
Robots can bake bread, cut carrots, and do karate kicks in demo videos. But how capable are they, really? @robocurve measures how good AI and robots are in the physical world, from making a sandwich to building data centers.
- X
Claude Fable 5 can control YAM arms through Inspect Robots to complete simple tasks:
Claude Fable 5 can control YAM arms through Inspect Robots to complete simple tasks:
- GitHub
inspect-robots
Official Robocurve repo for an open-source evaluation framework for VLA and physical-AI models.
- X
We are building @robocurve, a Public Benefit Corporation, to inform the public of the state of robotics progress. We are scaling the fiel...
We are building @robocurve, a Public Benefit Corporation, to inform the public of the state of robotics progress. We are scaling the field of robotics evals as fast as possible.
- X
We are an independent evaluator like METR, but for robots.
We are an independent evaluator like METR, but for robots.
- X
Today, we are releasing Inspect Robots, an open-source evaluation framework for AI and robots.
Today, we are releasing Inspect Robots, an open-source evaluation framework for AI and robots.
- X
Inspect Robots works with 40 VLAs (including π0.5, GR00T N1.7, MolmoAct2) and 260+ frontier LLMs on day one.
Inspect Robots works with 40 VLAs (including π0.5, GR00T N1.7, MolmoAct2) and 260+ frontier LLMs on day one.
- GitHub
Inspect Robots adapters for I2RT YAM bimanual arms driven by MolmoAct2
Inspect Robots adapters for I2RT YAM bimanual arms driven by MolmoAct2 Repository description: Inspect Robots adapters for I2RT YAM bimanual arms driven by MolmoAct2 The public non-fork repository is owned by the robocurve organization and was last pushed at...