BenchmarkPAW-GEN-10-Agentic
Environmente01Sort screws into bins
- Episodes
- Claude Fable 5.2 30 Claude Opus 5.5 30 DiT-Flow 30 Gemini 4 Argon 30 GPT-6 Astra 30 GR00T N1.7 30 MolmoAct 2 30 Pi 0.5 30 Rho 30
- Total duration
- Claude Fable 5.2 39:08 Claude Opus 5.5 51:44 DiT-Flow 27:40 Gemini 4 Argon 65:33 GPT-6 Astra 65:13 GR00T N1.7 23:20 MolmoAct 2 15:04 Pi 0.5 41:03 Rho 15:18
- Mean duration
- Claude Fable 5.2 78.3 s Claude Opus 5.5 103.5 s DiT-Flow 55.3 s Gemini 4 Argon 131.1 s GPT-6 Astra 130.4 s GR00T N1.7 46.7 s MolmoAct 2 30.1 s Pi 0.5 82.1 s Rho 30.6 s
- Success
- Claude Fable 5.2 83% Claude Opus 5.5 100% DiT-Flow 7% Gemini 4 Argon 80% GPT-6 Astra 77% GR00T N1.7 3% MolmoAct 2 0% Pi 0.5 3% Rho 7%
- Manipulation density
- Claude Fable 5.2 Share of time manipulating, five definitionsJoints moving87.4%End-effector moving67.0%Gripper engaged31.2%Contact force > 5 N55.4%Gripper engaged or in contact62.3% Claude Opus 5.5 Share of time manipulating, five definitionsJoints moving90.3%End-effector moving72.1%Gripper engaged60.9%Contact force > 5 N63.9%Gripper engaged or in contact86.1% DiT-Flow Share of time manipulating, five definitionsJoints moving90.4%End-effector moving80.5%Gripper engaged65.8%Contact force > 5 N83.2%Gripper engaged or in contact94.3% Gemini 4 Argon Share of time manipulating, five definitionsJoints moving72.0%End-effector moving53.7%Gripper engaged92.9%Contact force > 5 N80.8%Gripper engaged or in contact98.6% GPT-6 Astra Share of time manipulating, five definitionsJoints moving86.7%End-effector moving62.0%Gripper engaged38.0%Contact force > 5 N67.7%Gripper engaged or in contact78.2% GR00T N1.7 Share of time manipulating, five definitionsJoints moving99.8%End-effector moving97.0%Gripper engaged63.4%Contact force > 5 N84.7%Gripper engaged or in contact96.6% MolmoAct 2 Share of time manipulating, five definitionsJoints moving100.0%End-effector moving97.1%Gripper engaged68.5%Contact force > 5 N83.0%Gripper engaged or in contact94.2% Pi 0.5 Share of time manipulating, five definitionsJoints moving99.7%End-effector moving98.1%Gripper engaged53.8%Contact force > 5 N75.9%Gripper engaged or in contact86.7% Rho Share of time manipulating, five definitionsJoints moving99.9%End-effector moving93.9%Gripper engaged48.9%Contact force > 5 N82.5%Gripper engaged or in contact89.7%
- SPARC smoothness
- Claude Fable 5.2 Spectral arc length of the motionHow directly the arm moved, read from the shape of its speed spectrum. Unitless measure ≤ -1; the closer to -1, the smoother.The measure ignores how large a movement is, so a large and a small one compare directly. Shown here as the mean over this environment's measured episodes.Cutoff10 HzAmplitude threshold5%Zero-padding16×Signalend-effector translationBalasubramanian et al., On the analysis of movement smoothness, J Neuroeng Rehabil 2015. Claude Opus 5.5 Spectral arc length of the motionHow directly the arm moved, read from the shape of its speed spectrum. Unitless measure ≤ -1; the closer to -1, the smoother.The measure ignores how large a movement is, so a large and a small one compare directly. Shown here as the mean over this environment's measured episodes.Cutoff10 HzAmplitude threshold5%Zero-padding16×Signalend-effector translationBalasubramanian et al., On the analysis of movement smoothness, J Neuroeng Rehabil 2015. DiT-Flow Spectral arc length of the motionHow directly the arm moved, read from the shape of its speed spectrum. Unitless measure ≤ -1; the closer to -1, the smoother.The measure ignores how large a movement is, so a large and a small one compare directly. Shown here as the mean over this environment's measured episodes.Cutoff10 HzAmplitude threshold5%Zero-padding16×Signalend-effector translationBalasubramanian et al., On the analysis of movement smoothness, J Neuroeng Rehabil 2015. Gemini 4 Argon Spectral arc length of the motionHow directly the arm moved, read from the shape of its speed spectrum. Unitless measure ≤ -1; the closer to -1, the smoother.The measure ignores how large a movement is, so a large and a small one compare directly. Shown here as the mean over this environment's measured episodes.Cutoff10 HzAmplitude threshold5%Zero-padding16×Signalend-effector translationBalasubramanian et al., On the analysis of movement smoothness, J Neuroeng Rehabil 2015. GPT-6 Astra Spectral arc length of the motionHow directly the arm moved, read from the shape of its speed spectrum. Unitless measure ≤ -1; the closer to -1, the smoother.The measure ignores how large a movement is, so a large and a small one compare directly. Shown here as the mean over this environment's measured episodes.Cutoff10 HzAmplitude threshold5%Zero-padding16×Signalend-effector translationBalasubramanian et al., On the analysis of movement smoothness, J Neuroeng Rehabil 2015. GR00T N1.7 Spectral arc length of the motionHow directly the arm moved, read from the shape of its speed spectrum. Unitless measure ≤ -1; the closer to -1, the smoother.The measure ignores how large a movement is, so a large and a small one compare directly. Shown here as the mean over this environment's measured episodes.Cutoff10 HzAmplitude threshold5%Zero-padding16×Signalend-effector translationBalasubramanian et al., On the analysis of movement smoothness, J Neuroeng Rehabil 2015. MolmoAct 2 Spectral arc length of the motionHow directly the arm moved, read from the shape of its speed spectrum. Unitless measure ≤ -1; the closer to -1, the smoother.The measure ignores how large a movement is, so a large and a small one compare directly. Shown here as the mean over this environment's measured episodes.Cutoff10 HzAmplitude threshold5%Zero-padding16×Signalend-effector translationBalasubramanian et al., On the analysis of movement smoothness, J Neuroeng Rehabil 2015. Pi 0.5 Spectral arc length of the motionHow directly the arm moved, read from the shape of its speed spectrum. Unitless measure ≤ -1; the closer to -1, the smoother.The measure ignores how large a movement is, so a large and a small one compare directly. Shown here as the mean over this environment's measured episodes.Cutoff10 HzAmplitude threshold5%Zero-padding16×Signalend-effector translationBalasubramanian et al., On the analysis of movement smoothness, J Neuroeng Rehabil 2015. Rho Spectral arc length of the motionHow directly the arm moved, read from the shape of its speed spectrum. Unitless measure ≤ -1; the closer to -1, the smoother.The measure ignores how large a movement is, so a large and a small one compare directly. Shown here as the mean over this environment's measured episodes.Cutoff10 HzAmplitude threshold5%Zero-padding16×Signalend-effector translationBalasubramanian et al., On the analysis of movement smoothness, J Neuroeng Rehabil 2015.
Episode
Loading the replay…
0:001:23