RoboCasa365
Can the model plan, use tools, and finish a multi-step task?
Agents5 ranked models6 reported values1 reportsHigher is better
Top rankings
One row per model, using its best reported score across effort settings.
| Rank | Model | Best score | Best reported setting |
|---|---|---|---|
| 1 | Qwen-RobotManipQwen | 68.6% | Reported configurationQwen report2 values · 1 reportJun 15, 2026 · source |
| 2 | RLDX-1External baseline | 63.0% | Reported configurationQwen report1 value · 1 reportJun 15, 2026 · source |
| 3 | GR00T-N1.5NVIDIA | 50.7% | Reported configurationQwen report1 value · 1 reportJun 15, 2026 · source |
| 4 | π₀.₅Physical Intelligence | 39.6% | Reported configurationQwen report1 value · 1 reportJun 15, 2026 · source |
| 5 | π₀Physical Intelligence | 36.3% | Reported configurationQwen report1 value · 1 reportJun 15, 2026 · source |
Effort curve
Every sourced cost-linked effort value for this exact version. Lines connect complete sweeps only.
No cost-linked effort sweep for this version.
Definition and comparison boundarypublic methodology
Can the model plan, use tools, and finish a multi-step task?
Agentic execution across tools, environments, or long-running workflows. Higher is better. The value is the percentage reported in this lab's table.
Harness, tools, prompts, attempt count, and benchmark version can materially change the result. Compare within one reporting context.
RoboCasa365 atomic and long-horizon composite success rates; Composite-Unseen is the held-out long-horizon split.
Method / source