Categories Companies Research

Loading the SOTA2 catalog…

Explore

Categories Companies Research

Contribute

Join SOTA2 Contact

SOTA2

© 2026 SOTA2

Build in Bellevue, USA

WizWand is now SOTA2 Research

Papers Benchmarks Datasets Tasks

Robot Manipulation research benchmarks – Page 12 · SOTA2 Research

Robot Manipulation

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

Compositional Planning Bench Puzzle IND

Compositional Visual Planning via Inference-Time Diffusion Scaling

50Success Rate

7

Mar 4, 2026

Compositional Planning Bench Cube, OOD

Compositional Visual Planning via Inference-Time Diffusion Scaling

65Success Rate

7

Mar 4, 2026

Compositional Planning Bench Cube, IND

Compositional Visual Planning via Inference-Time Diffusion Scaling

64Success Rate

7

Mar 4, 2026

Compositional Planning Bench Drawer OOD

Compositional Visual Planning via Inference-Time Diffusion Scaling

52Success Rate

7

Mar 4, 2026

Compositional Planning Bench Drawer, IND

54Success Rate

7

Mar 4, 2026

Compositional Planning Bench Tool-Use OOD

Compositional Visual Planning via Inference-Time Diffusion Scaling

96Success Rate

7

Mar 4, 2026

Compositional Planning Bench Tool-Use IND

Compositional Visual Planning via Inference-Time Diffusion Scaling

97Success Rate

7

Mar 4, 2026

MemoryBench (test)

100Success Rate

7

Feb 26, 2026

0.96Success Rate (Baseline)

7

Feb 26, 2026

Real-world manipulation tasks (Unseen Tasks)

74Success Rate

7

Feb 26, 2026

RoboCasa simulation

55.6Average Success Rate

7

Mar 18, 2026

RLBench Simulation (test)

0.68Avg ATP

7

Feb 26, 2026

SimplerEnv Google Robot tasks - Overall

71.8Average Success

7

Feb 26, 2026

Real-world post-training dataset Task 2: Move condiment cup into slot 1.0 (test)

86.7Success Rate

7

Feb 26, 2026

Real-world post-training dataset Task 1: Move pink tulip to vase 1.0 (test)

93.3Success Rate

7

Feb 26, 2026

96CloseBox Success Rate

7

Feb 26, 2026

Meta-world v1 v2 (train test)

94.4Basketball

7

Feb 26, 2026

Unseen Household OOD Shift New (test)

56Transfer Bottle

6

Jul 9, 2026

Standardized Robot Cubicle Orig. in-distribution (train)

77.8Transfer Bottle Success Rate

6

Jul 9, 2026

Unseen Home Environments

56Transfer Bottle Success Rate

6

Jul 9, 2026

RoboTwin Hard 2.0

45Click Alarm Success Rate

6

Jul 7, 2026

RoboTwin Easy 2.0

96Click Alarm Success Rate

6

Jul 7, 2026

Libero 10 (held-out conditions)

75Success Rate

6

Jul 3, 2026

Push-T (held-out conditions)

65Success Rate

6

Jul 3, 2026

Robomimic Tool Hang (held-out conditions)

39Success Rate

6

Jul 3, 2026

Page 12 of 31

Popular datasets

LIBERO RoboTwin RLBench SimplerEnv Real-world Robot Manipulation

Follow for updates

Join SOTA2 Contact