Manipulation benchmark
Can the model plan, use tools, and finish a multi-step task?
Agents2 ranked models2 reported values1 reportsHigher is better
Top rankings
One row per model, using its best reported score across effort settings.
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.
Qwen-VLA-Instruct unified manipulation table; specialist baseline cells are retained exactly where printed.
Method / source