# Robot Manipulation on RoboCasa-GR1 24 tasks Source: [Robot Manipulation on RoboCasa-GR1 24 tasks]() Collection index: [llms.txt]() Data updated: 2026-04-30T07:55:19.085Z Evaluation of robot manipulation performance across the 24 specific tasks defined within the RoboCasa GR1 benchmark suite. A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Robot Manipulation | | Dataset (including any stated split) | RoboCasa-GR1 24 tasks | | Metric | Average Success Rate | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 16 of 16 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [Average Success Rate](): Higher is better; unit: PERCENT. ## Recorded results | Method | Average Success Rate | Result date | Report paper | | --- | --- | --- | --- | | X-WAM | 79.2 | Not recorded | 69f2e20955a3c32af89fc5f6 | | Cosmos Policy | 67.1 | Not recorded | 69f2e20955a3c32af89fc5f6 | | GR00T-N1.5 | 64.1 | Not recorded | 69f2e20955a3c32af89fc5f6 | | π0 | 62.5 | Not recorded | 69f2e20955a3c32af89fc5f6 | | DreamZero | 62.4 | Not recorded | 69f2e20955a3c32af89fc5f6 | | UWM | 60.8 | Not recorded | 69f2e20955a3c32af89fc5f6 | | StarVLA-α | 57.3 | Not recorded | 69ddca837d0172140ec7a01c | | StarVLA-α | 53.8 | Not recorded | 69ddca837d0172140ec7a01c | | StarVLA-OFT | 48.8 | Not recorded | 69d5e11f7d0172140ec75af3 | | StarVLA-GR00T | 47.8 | Not recorded | 69d5e11f7d0172140ec75af3 | | GR00T-N1.6 | 47.6 | Not recorded | 69d5e11f7d0172140ec75af3 | | GR00T-N1.6 | 47.6 | Not recorded | 69ddca837d0172140ec7a01c | | StarVLA-π | 43.9 | Not recorded | 69d5e11f7d0172140ec75af3 | | StarVLA-FAST | 39 | Not recorded | 69d5e11f7d0172140ec75af3 | | π0.5 | 37 | Not recorded | 69d5e11f7d0172140ec75af3 | | π0.5 | 37 | Not recorded | 69ddca837d0172140ec7a01c | ### X-WAM Average Success Rate: 79.2. Result ID: 69f30ad947f34ce924d93ef5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"WAM-based methods"}. This record does not assert independent reproduction. ### Cosmos Policy Average Success Rate: 67.1. Result ID: 69f30ad947f34ce924d93ef4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"WAM-based methods"}. This record does not assert independent reproduction. ### GR00T-N1.5 Average Success Rate: 64.1. Result ID: 69f30ad947f34ce924d93ef1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"VLA-based methods"}. This record does not assert independent reproduction. ### π0 Average Success Rate: 62.5. Result ID: 69f30ad947f34ce924d93ef0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"VLA-based methods"}. This record does not assert independent reproduction. ### DreamZero Average Success Rate: 62.4. Result ID: 69f30ad947f34ce924d93ef3. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Category":"WAM-based methods"}. This record does not assert independent reproduction. ### UWM Average Success Rate: 60.8. Result ID: 69f30ad947f34ce924d93ef2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"WAM-based methods"}. This record does not assert independent reproduction. ### StarVLA-α Average Success Rate: 57.3. Result ID: 69de343d47f34ce924c71417. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Protocol":"Generalist"}. This record does not assert independent reproduction. ### StarVLA-α Average Success Rate: 53.8. Result ID: 69de343d47f34ce924c71416. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Protocol":"Specialist"}. This record does not assert independent reproduction. ### StarVLA-OFT Average Success Rate: 48.8. Result ID: 69d606ea47f34ce924be4845. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training setting":"single-benchmark"}. This record does not assert independent reproduction. ### StarVLA-GR00T Average Success Rate: 47.8. Result ID: 69d606ea47f34ce924be4842. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training setting":"single-benchmark"}. This record does not assert independent reproduction. ### GR00T-N1.6 Average Success Rate: 47.6. Result ID: 69d606ea47f34ce924be4843. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training setting":"single-benchmark"}. This record does not assert independent reproduction. ### GR00T-N1.6 Average Success Rate: 47.6. Result ID: 69de343d47f34ce924c71415. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Protocol":"Specialist"}. This record does not assert independent reproduction. ### StarVLA-π Average Success Rate: 43.9. Result ID: 69d606ea47f34ce924be4844. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training setting":"single-benchmark"}. This record does not assert independent reproduction. ### StarVLA-FAST Average Success Rate: 39. Result ID: 69d606ea47f34ce924be4841. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training setting":"single-benchmark"}. This record does not assert independent reproduction. ### π0.5 Average Success Rate: 37. Result ID: 69d606ea47f34ce924be4840. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training setting":"single-benchmark"}. This record does not assert independent reproduction. ### π0.5 Average Success Rate: 37. Result ID: 69de343d47f34ce924c71414. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Protocol":"Specialist"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.