BenchmarkPAW-GEN-10-Agentic
Environmente03Pour screws into the funnel
- Episodes
- GPT-6 Astra 30 Claude Fable 5.2 30 Claude Opus 5.5 30 DiT-Flow 30 Gemini 4 Argon 30 GR00T N1.7 30 MolmoAct 2 30 Pi 0.5 30 Rho 30
- Total duration
- GPT-6 Astra 29:21 Claude Fable 5.2 27:00 Claude Opus 5.5 26:03 DiT-Flow 6:18 Gemini 4 Argon 19:45 GR00T N1.7 4:36 MolmoAct 2 6:40 Pi 0.5 6:36 Rho 3:22
- Mean duration
- GPT-6 Astra 58.7 s Claude Fable 5.2 54.0 s Claude Opus 5.5 52.1 s DiT-Flow 12.6 s Gemini 4 Argon 39.5 s GR00T N1.7 9.2 s MolmoAct 2 13.3 s Pi 0.5 13.2 s Rho 6.7 s
- Success
- GPT-6 Astra 73% Claude Fable 5.2 87% Claude Opus 5.5 97% DiT-Flow 23% Gemini 4 Argon 100% GR00T N1.7 7% MolmoAct 2 30% Pi 0.5 13% Rho 13%
- Manipulation density
- GPT-6 Astra Share of time manipulating, five definitionsJoints moving81.2%End-effector moving62.8%Gripper engaged55.9%Contact force > 5 N75.0%Gripper engaged or in contact80.9% Claude Fable 5.2 Share of time manipulating, five definitionsJoints moving85.7%End-effector moving64.8%Gripper engaged53.7%Contact force > 5 N39.4%Gripper engaged or in contact71.7% Claude Opus 5.5 Share of time manipulating, five definitionsJoints moving94.1%End-effector moving84.2%Gripper engaged65.3%Contact force > 5 N62.1%Gripper engaged or in contact81.6% DiT-Flow Share of time manipulating, five definitionsJoints moving100.0%End-effector moving98.7%Gripper engaged84.8%Contact force > 5 N48.8%Gripper engaged or in contact92.0% Gemini 4 Argon Share of time manipulating, five definitionsJoints moving90.0%End-effector moving80.6%Gripper engaged63.9%Contact force > 5 N81.7%Gripper engaged or in contact90.5% GR00T N1.7 Share of time manipulating, five definitionsJoints moving100.0%End-effector moving99.6%Gripper engaged72.1%Contact force > 5 N63.1%Gripper engaged or in contact86.2% MolmoAct 2 Share of time manipulating, five definitionsJoints moving100.0%End-effector moving99.4%Gripper engaged71.3%Contact force > 5 N53.1%Gripper engaged or in contact81.8% Pi 0.5 Share of time manipulating, five definitionsJoints moving99.9%End-effector moving99.6%Gripper engaged82.5%Contact force > 5 N45.3%Gripper engaged or in contact89.4% Rho Share of time manipulating, five definitionsJoints moving100.0%End-effector moving99.3%Gripper engaged70.6%Contact force > 5 N61.1%Gripper engaged or in contact82.3%
- SPARC smoothness
- 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. 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. 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…
0:001:07