{ "$schema": "./schema/leaderboard.schema.json", "version": 76, "generated_at": "2026-10-23T22:46:44Z", "generator": "build_leaderboard.py", "benchmark": { "name": "RoboLab-120", "task_count": 120, "episodes_per_task": 10, "robot": "DROID (Franka)", "difficulty_tiers": { "simple": 64, "moderate": 39, "complex": 17 }, "metrics": { "sr": "success rate (%), 95% Clopper-Pearson CI", "score": "subtask-progress score x100, 95% task-level cluster bootstrap CI", "score_ci_bootstrap": "per-slice 95% task-level cluster bootstrap CI on the score (published on the website)", "ee_speed": "end-effector speed, cm/s, excludes inference time", "ee_speed_sd": "population SD of the per-episode end-effector speed, cm/s (ee_speed_sd_cm_s)", "ee_sparc": "end-effector spectral arc length (closer to 0 is smoother)" }, "competency_axes": { "procedural": { "attributes": [ "affordance", "reorientation", "stacking" ], "task_count": 22 }, "relational": { "attributes": [ "conjunction", "counting", "spatial" ], "task_count": 42 }, "visual": { "attributes": [ "color", "semantics", "size" ], "task_count": 84 } } }, "entries": [ { "id": "sensenova-roborsi", "display_name": "SenseNova-RoboRSI", "organization": "Nanyang Technological University", "organization_links": [], "source": "external-submission", "badges": [ "NEW" ], "policy": { "type": "other", "type_other": "Agentic LLM", "description": "GPT-6-Astra agent implemented in SenseNova-RoboRSI, with GPT-6-Astra directly acting as the policy and no additional learned model or benchmark-specific adaptation.", "size": "--", "vram": "--", "robot": "DROID", "cameras": [ "EGO-L", "Wrist" ], "observation_space": [ "RGB Image", "Proprioception", "Natural-language task instruction" ], "observation_history_label": "At observation step t (1-indexed), up to t−1 prior steps of two-view RGB and proprioception are retained within the episode, subject to context limits.", "action_space": [ "End effector pose" ], "action_horizon_steps": null, "gpu": "NVIDIA RTX 4090" }, "open_sourcing": { "status": "open" }, "training": { "sim_image_data": { "robolab_in_finetuning": false, "robolab_in_training": null, "other_sim_in_training": null }, "simulation_use": "No additional training or fine-tuning; the foundation model’s pretraining data and simulation-data proportion are undisclosed" }, "provenance": { "robolab_build": "92ccc0ee1984a272d4748c92ed73144fa806a6f4 (v0.3.1 + 7 fork commits)", "isaac_versions": "0.54.2 / 6.0.0.1", "display_date": "Submitted Oct 18, 2026" }, "results": { "default": { "episodes": 1200, "successes": 685, "sr": 57.1, "score": 69.9, "score_ci_bootstrap": [ 64.4, 75.1 ], "ee_speed_cm_s": 5.48, "ee_speed_sd_cm_s": 2.57, "ee_sparc": -4.04, "by_difficulty": { "simple": { "episodes": 640, "successes": 380, "sr": 59.4, "score": 67.9, "score_ci_bootstrap": [ 60.0, 75.3 ], "sr_ci_cp": [ 55.45581711870105, 63.20730182928539 ] }, "moderate": { "episodes": 390, "successes": 219, "sr": 56.2, "score": 69.1, "score_ci_bootstrap": [ 59.7, 78.1 ], "sr_ci_cp": [ 51.07006428222576, 61.143531437431385 ] }, "complex": { "episodes": 170, "successes": 86, "sr": 50.6, "score": 79.3, "score_ci_bootstrap": [ 68.9, 88.5 ], "sr_ci_cp": [ 42.82630753719037, 58.329297405135506 ] } }, "by_axis": { "procedural": { "episodes": 220, "successes": 119, "sr": 54.1, "score": 66.1, "score_ci_bootstrap": [ 54.8, 77.1 ], "sr_ci_cp": [ 47.26182593697879, 60.808265582339935 ] }, "relational": { "episodes": 420, "successes": 233, "sr": 55.5, "score": 64.2, "score_ci_bootstrap": [ 53.9, 74.0 ], "sr_ci_cp": [ 50.57964818574449, 60.295053761070314 ] }, "visual": { "episodes": 840, "successes": 421, "sr": 50.1, "score": 67.1, "score_ci_bootstrap": [ 60.6, 73.3 ], "sr_ci_cp": [ 46.682927586747994, 53.55432973523988 ] } }, "episodes_per_task": "10", "sr_ci_cp": [ 54.22672820856893, 59.90514231706155 ] } }, "links": [ { "label": "Paper", "url": "https://arxiv.org/abs/2610.22471", "kind": "paper" }, { "label": "Project page", "url": "https://sensenova.ai/has-gpt-6-astra-achieved-embodied-intelligence", "kind": "project" } ], "table": { "observations": [ "RGB", "P", "L" ] } }, { "id": "lakeside", "display_name": "Lakeside", "organization": "PrI", "organization_links": [], "source": "external-submission", "policy": { "type": "other", "type_other": "LWM", "description": "Latent World Model (LWM)", "robot": "DROID (Franka)", "observation_space": [ "RGB", "Proprioception" ], "action_space": [ "Joint position" ], "action_horizon_steps": "32 steps", "size": "5B", "gpu": "RTX 5090", "vram": "27.2 GB", "cameras": [ "EGO-L/R", "Wrist" ], "observation_history_label": "Current" }, "open_sourcing": { "status": "closed" }, "training": { "sim_image_data": { "robolab_in_finetuning": false, "robolab_in_training": false, "other_sim_in_training": true }, "simulation_use": "Other simulation data (1.6%)" }, "provenance": { "display_date": "Submitted Sep 16, 2026", "robolab_build": "0aef241f (v0.2.1)", "isaac_versions": "2.2.0 / 5.0.0" }, "results": { "default": { "episodes": 3000, "successes": 1531, "sr": 51.0, "score": 65.3, "score_ci_bootstrap": [ 59.0, 71.4 ], "ee_speed_cm_s": 7.1, "ee_speed_sd_cm_s": 2.06, "ee_sparc": -4.84, "by_difficulty": { "simple": { "episodes": 1600, "successes": 921, "sr": 57.6, "score": 65.2, "score_ci_bootstrap": [ 56.2, 73.8 ], "sr_ci_cp": [ 55.09744486139488, 59.99975474904389 ] }, "moderate": { "episodes": 975, "successes": 430, "sr": 44.1, "score": 63.5, "score_ci_bootstrap": [ 52.1, 74.2 ], "sr_ci_cp": [ 40.9565828468055, 47.284263290683654 ] }, "complex": { "episodes": 425, "successes": 180, "sr": 42.4, "score": 69.5, "score_ci_bootstrap": [ 54.7, 82.7 ], "sr_ci_cp": [ 37.60518966737408, 47.20789294667248 ] } }, "by_axis": { "procedural": { "episodes": 550, "successes": 169, "sr": 30.7, "score": 50.2, "score_ci_bootstrap": [ 37.0, 63.5 ], "sr_ci_cp": [ 26.893039864044244, 34.770056550089876 ] }, "relational": { "episodes": 1050, "successes": 564, "sr": 53.7, "score": 67.4, "score_ci_bootstrap": [ 54.9, 79.1 ], "sr_ci_cp": [ 50.64389076430491, 56.763807361114175 ] }, "visual": { "episodes": 2100, "successes": 993, "sr": 47.3, "score": 61.8, "score_ci_bootstrap": [ 54.6, 68.8 ], "sr_ci_cp": [ 45.131616545927, 49.4473987404347 ] } }, "episodes_per_task": "25", "sr_ci_cp": [ 49.227583371233834, 52.837065878249504 ] } }, "badges": [ "Organizer cross-check" ], "links": [], "table": { "observations": [ "RGB", "P" ] } }, { "id": "bimind-v0-5", "display_name": "BiMind v0.5", "organization": "Bicameral Robotics", "organization_links": [ { "label": "Bicameral Robotics", "url": "https://bicameralrobotics.ai/" } ], "source": "external-submission", "policy": { "type": "VLA", "type_other": null, "description": "A VLA with long horizon memory", "robot": "DROID", "observation_space": [ "rgb", "proprioception" ], "observation_history_label": "all", "action_space": [ "joint position" ], "action_horizon_steps": null, "size": "8B", "gpu": "NVIDIA RTX 5090", "vram": "30 GiB" }, "open_sourcing": { "status": "closed" }, "training": { "sim_image_data": { "robolab_in_finetuning": true, "robolab_in_training": false, "other_sim_in_training": true }, "simulation_use": "RoboLab/IsaacLab/IsaacSim data in fine-tuning; Other simulation data. Simulation comprises approximately 10% of all samples across pretraining and fine-tuning." }, "provenance": { "robolab_build": "7d45d74904eade3b578a8eb1f2f9f89bc3d40326 (v0.1.1)", "isaac_versions": "2.2.0 / 5.0.0", "display_date": "Submitted Oct 13, 2026" }, "results": { "default": { "episodes": 1200, "successes": 590, "sr": 49.2, "score": null, "score_ci_bootstrap": null, "ee_speed_cm_s": 5.42, "ee_speed_sd_cm_s": 1.96, "ee_sparc": -5.07, "by_difficulty": { "simple": { "episodes": 640, "successes": 330, "sr": 51.6, "score": null, "score_ci_bootstrap": null, "sr_ci_cp": [ 47.612973204418694, 55.49755307886768 ] }, "moderate": { "episodes": 390, "successes": 205, "sr": 52.6, "score": null, "score_ci_bootstrap": null, "sr_ci_cp": [ 47.476771243607445, 57.612234937373906 ] }, "complex": { "episodes": 170, "successes": 55, "sr": 32.4, "score": null, "score_ci_bootstrap": null, "sr_ci_cp": [ 25.391421423232647, 39.942431685884685 ] } }, "by_axis": { "procedural": { "episodes": 220, "successes": 96, "sr": 43.6, "score": null, "score_ci_bootstrap": null, "sr_ci_cp": [ 36.98127974227961, 50.46527949871348 ] }, "relational": { "episodes": 420, "successes": 246, "sr": 58.6, "score": null, "score_ci_bootstrap": null, "sr_ci_cp": [ 53.69468613277439, 63.32655272723691 ] }, "visual": { "episodes": 840, "successes": 342, "sr": 40.7, "score": null, "score_ci_bootstrap": null, "sr_ci_cp": [ 37.36991171878735, 44.12404299972232 ] } }, "episodes_per_task": "10", "sr_ci_cp": [ 46.301426141682114, 52.036000080833254 ] }, "vague": { "episodes": 1200, "successes": 555, "sr": 46.2, "score": null, "score_ci_bootstrap": null, "ee_speed_cm_s": 5.37, "ee_speed_sd_cm_s": 1.96, "ee_sparc": -5.13, "by_difficulty": { "simple": { "episodes": 640, "successes": 315, "sr": 49.2, "score": null, "score_ci_bootstrap": null, "sr_ci_cp": [ 45.27866497530632, 53.16607181565144 ] }, "moderate": { "episodes": 390, "successes": 197, "sr": 50.5, "score": null, "score_ci_bootstrap": null, "sr_ci_cp": [ 45.43494071703938, 55.58286083682986 ] }, "complex": { "episodes": 170, "successes": 43, "sr": 25.3, "score": null, "score_ci_bootstrap": null, "sr_ci_cp": [ 18.950087401633148, 32.520610481148104 ] } }, "by_axis": { "procedural": { "episodes": 220, "successes": 84, "sr": 38.2, "score": null, "score_ci_bootstrap": null, "sr_ci_cp": [ 31.733331511420726, 44.95341675967789 ] }, "relational": { "episodes": 420, "successes": 236, "sr": 56.2, "score": null, "score_ci_bootstrap": null, "sr_ci_cp": [ 51.29684236834016, 60.99629434034357 ] }, "visual": { "episodes": 840, "successes": 318, "sr": 37.9, "score": null, "score_ci_bootstrap": null, "sr_ci_cp": [ 34.56511920801814, 41.23469283303652 ] } }, "episodes_per_task": "10", "sr_ci_cp": [ 43.39948141588907, 49.11893751682276 ] }, "specific": { "episodes": 1200, "successes": 567, "sr": 47.2, "score": null, "score_ci_bootstrap": null, "ee_speed_cm_s": 5.38, "ee_speed_sd_cm_s": 1.96, "ee_sparc": -5.15, "by_difficulty": { "simple": { "episodes": 640, "successes": 323, "sr": 50.5, "score": null, "score_ci_bootstrap": null, "sr_ci_cp": [ 46.52257407070307, 54.4105838633429 ] }, "moderate": { "episodes": 390, "successes": 191, "sr": 49.0, "score": null, "score_ci_bootstrap": null, "sr_ci_cp": [ 43.90901652359251, 54.05538044795757 ] }, "complex": { "episodes": 170, "successes": 53, "sr": 31.2, "score": null, "score_ci_bootstrap": null, "sr_ci_cp": [ 24.304271397449963, 38.718834696433184 ] } }, "by_axis": { "procedural": { "episodes": 220, "successes": 77, "sr": 35.0, "score": null, "score_ci_bootstrap": null, "sr_ci_cp": [ 28.71093232284872, 41.69949635632291 ] }, "relational": { "episodes": 420, "successes": 228, "sr": 54.3, "score": null, "score_ci_bootstrap": null, "sr_ci_cp": [ 49.38651583614428, 59.12412490630531 ] }, "visual": { "episodes": 840, "successes": 335, "sr": 39.9, "score": null, "score_ci_bootstrap": null, "sr_ci_cp": [ 36.55062914022278, 43.282532090312415 ] } }, "episodes_per_task": "10", "sr_ci_cp": [ 44.39334746097927, 50.120159423552856 ] } }, "badges": [ "NEW" ], "links": [ { "label": "Project page", "url": "https://bicameralrobotics.ai/", "kind": "project" } ] }, { "id": "flux3-action", "display_name": "FLUX 3 Action", "organization": "Black Forest Labs", "organization_links": [ { "label": "Black Forest Labs", "url": "https://bfl.ai/" } ], "source": "external-submission", "policy": { "type": "WAM", "type_other": null, "description": "Built from the FLUX.3 Action open-weights base model and fine-tuned on DROID demonstrations.", "robot": "DROID (Franka + Robotiq 2F-85)", "observation_space": [ "RGB", "Proprioception" ], "action_space": [ "Joint position" ], "action_horizon_steps": "32 steps", "size": "7B", "gpu": "Heterogeneous", "vram": "69 GiB", "cameras": [ "EGO-L/R", "Wrist" ], "observation_history_label": "Current" }, "open_sourcing": { "status": "open" }, "training": { "sim_image_data": { "robolab_in_finetuning": false, "robolab_in_training": false, "other_sim_in_training": false }, "simulation_use": null }, "provenance": { "display_date": "Submitted Sep 15, 2026", "robolab_build": "38d0b1c0", "isaac_versions": "2.2.0 / 5.0.0" }, "results": { "default": { "episodes": 1200, "successes": 515, "sr": 42.9, "score": null, "score_ci_bootstrap": null, "ee_speed_cm_s": 7.36, "ee_speed_sd_cm_s": 1.85, "ee_sparc": -5.37, "by_difficulty": { "simple": { "episodes": 640, "successes": 314, "sr": 49.1, "score": null, "score_ci_bootstrap": null, "sr_ci_cp": [ 45.123345938820506, 53.01033822218555 ] }, "moderate": { "episodes": 390, "successes": 153, "sr": 39.2, "score": null, "score_ci_bootstrap": null, "sr_ci_cp": [ 34.35464631906901, 44.2716474739378 ] }, "complex": { "episodes": 170, "successes": 48, "sr": 28.2, "score": null, "score_ci_bootstrap": null, "sr_ci_cp": [ 21.609468343462616, 35.63711916012338 ] } }, "by_axis": { "procedural": { "episodes": 220, "successes": 79, "sr": 35.9, "score": null, "score_ci_bootstrap": null, "sr_ci_cp": [ 29.571433156111077, 42.632206446285906 ] }, "relational": { "episodes": 420, "successes": 208, "sr": 49.5, "score": null, "score_ci_bootstrap": null, "sr_ci_cp": [ 44.64110382122211, 54.413268666983214 ] }, "visual": { "episodes": 840, "successes": 299, "sr": 35.6, "score": null, "score_ci_bootstrap": null, "sr_ci_cp": [ 32.35319877204357, 38.938765466254 ] } }, "episodes_per_task": "10", "sr_ci_cp": [ 40.09485768293845, 45.77327179143107 ] } }, "badges": [ "Organizer cross-check" ], "links": [ { "label": "Project Page", "url": "https://bfl.ai/models/flux-3-action", "kind": "project" }, { "label": "Model Weights", "url": "https://huggingface.co/collections/black-forest-labs/flux-3-action", "kind": "model" }, { "label": "Code", "url": "https://github.com/black-forest-labs/flux-action", "kind": "code" } ], "table": { "observations": [ "RGB", "P" ] } }, { "id": "cosmic", "display_name": "Cosmic", "organization": "Geodesic", "organization_links": [ { "label": "Geodesic", "url": "https://www.geodesicos.com/" } ], "source": "external-submission", "policy": { "type": "other", "type_other": "TAMP+FM", "description": "TAMP+FM manipulation stack with no fine-tuning on DROID demonstrations", "robot": "DROID (Franka + Robotiq 2F-85)", "observation_space": [ "RGB", "Depth", "Proprioception" ], "action_space": [ "Joint position" ], "action_horizon_steps": null, "size": "~350M", "gpu": "NVIDIA RTX 4090", "vram": "10 GB", "observation_history_label": "Current", "cameras": null }, "open_sourcing": { "status": "closed" }, "training": { "sim_image_data": { "robolab_in_finetuning": false, "robolab_in_training": false, "other_sim_in_training": true } }, "provenance": { "display_date": "Submitted Oct 3, 2026", "robolab_build": "0aef241f (v0.2.1)", "isaac_versions": "2.2 / 5.0" }, "results": { "default": { "episodes": 1200, "successes": 508, "sr": 42.3, "score": 59.7, "score_ci_bootstrap": [ 54.0, 65.3 ], "ee_speed_cm_s": 7.87, "ee_speed_sd_cm_s": 3.02, "ee_sparc": -7.2, "by_difficulty": { "simple": { "episodes": 640, "successes": 298, "sr": 46.6, "score": 57.2, "score_ci_bootstrap": [ 49.1, 65.0 ], "sr_ci_cp": [ 42.64337706065518, 50.51346663955657 ] }, "moderate": { "episodes": 390, "successes": 156, "sr": 40.0, "score": 59.8, "score_ci_bootstrap": [ 49.7, 69.6 ], "sr_ci_cp": [ 35.1018796913258, 45.05109069684462 ] }, "complex": { "episodes": 170, "successes": 54, "sr": 31.8, "score": 69.0, "score_ci_bootstrap": [ 58.5, 79.6 ], "sr_ci_cp": [ 24.847208123795756, 39.33126260208092 ] } }, "by_axis": { "procedural": { "episodes": 220, "successes": 71, "sr": 32.3, "score": 51.0, "score_ci_bootstrap": [ 37.8, 64.2 ], "sr_ci_cp": [ 26.144602580963795, 38.88632926738304 ] }, "relational": { "episodes": 420, "successes": 164, "sr": 39.0, "score": 54.9, "score_ci_bootstrap": [ 44.3, 65.2 ], "sr_ci_cp": [ 34.354480531556156, 43.89620609530327 ] }, "visual": { "episodes": 840, "successes": 334, "sr": 39.8, "score": 58.2, "score_ci_bootstrap": [ 51.3, 65.1 ], "sr_ci_cp": [ 36.43366997641794, 43.16223515820391 ] } }, "episodes_per_task": "10", "sr_ci_cp": [ 39.51786952868454, 45.18646010450809 ] } }, "badges": [ "NEW" ], "links": [ { "label": "Project", "url": "https://www.geodesicos.com/technology.html", "kind": "project" } ], "table": { "observations": [ "RGB", "D", "P" ] } }, { "id": "hidream-o1-embodied", "display_name": "HiDream-O1-Embodied", "organization": "HiDream.ai", "organization_links": [ { "label": "HiDream.ai", "url": "https://www.hidreamai.com/home" } ], "source": "external-submission", "policy": { "type": "VLA", "type_other": null, "description": "Based on HiDream-O1’s omni-modal technology framework, enhancing robots’ ability to predict future events.", "robot": "DROID (Franka)", "observation_space": [ "RGB", "Proprioception" ], "action_space": [ "Joint position" ], "action_horizon_steps": null, "size": "6B", "gpu": "8× NVIDIA A100 80GB", "vram": "16 GB", "cameras": [ "EGO-L/R", "Wrist" ], "observation_history_label": "Current" }, "open_sourcing": { "status": "closed" }, "training": { "sim_image_data": { "robolab_in_finetuning": true, "robolab_in_training": false, "other_sim_in_training": false }, "simulation_use": "RoboLab sim in fine-tuning (3.7%)" }, "provenance": { "display_date": "Submitted Sep 16, 2026", "robolab_build": "9db0aaf0 (v0.3.1)", "isaac_versions": "2.2.0 / 5.0.0" }, "results": { "default": { "episodes": 1200, "successes": 479, "sr": 39.9, "score": 50.5, "score_ci_bootstrap": [ 43.0, 57.7 ], "ee_speed_cm_s": 5.32, "ee_speed_sd_cm_s": 2.23, "ee_sparc": -4.95, "by_difficulty": { "simple": { "episodes": 640, "successes": 276, "sr": 43.1, "score": 47.7, "score_ci_bootstrap": [ 37.1, 58.3 ], "sr_ci_cp": [ 39.249324362365066, 47.06437471504338 ] }, "moderate": { "episodes": 390, "successes": 147, "sr": 37.7, "score": 54.3, "score_ci_bootstrap": [ 41.9, 66.5 ], "sr_ci_cp": [ 32.86395860330415, 42.70899501711962 ] }, "complex": { "episodes": 170, "successes": 56, "sr": 32.9, "score": 52.2, "score_ci_bootstrap": [ 35.5, 69.2 ], "sr_ci_cp": [ 25.936892667972426, 40.55235960182142 ] } }, "by_axis": { "procedural": { "episodes": 220, "successes": 65, "sr": 29.5, "score": 46.7, "score_ci_bootstrap": [ 31.8, 61.4 ], "sr_ci_cp": [ 23.602176377292675, 36.049509818721205 ] }, "relational": { "episodes": 420, "successes": 202, "sr": 48.1, "score": 56.4, "score_ci_bootstrap": [ 43.2, 69.2 ], "sr_ci_cp": [ 43.22589766967922, 52.991592852611745 ] }, "visual": { "episodes": 840, "successes": 224, "sr": 26.7, "score": 39.1, "score_ci_bootstrap": [ 30.9, 47.3 ], "sr_ci_cp": [ 23.702868250077515, 29.79517515501733 ] } }, "episodes_per_task": "10", "sr_ci_cp": [ 37.13176058113532, 42.75111614052884 ] } }, "badges": [ "Organizer cross-check" ], "links": [], "table": { "observations": [ "RGB", "P" ] } }, { "id": "atomic-wam", "display_name": "Atomic-WAM", "organization": "ZJU-APRIL & ACE Robotics", "organization_links": [ { "label": "ZJU-APRIL", "url": "https://april.zju.edu.cn/" }, { "label": "ACE Robotics", "url": "https://www.acerobotics.com/" } ], "source": "external-submission", "policy": { "type": "VLM+WAM", "type_other": null, "description": "Atomic-WAM is an understand-then-execute framework that connects VLM task-reasoning to WAM action generation through atomic-motion descriptions.", "robot": "DROID (Franka)", "observation_space": [ "RGB", "Proprioception" ], "observation_history_label": "VLM: 3-frame RGB; WAM: current RGB frame plus proprio", "action_space": [ "Joint position" ], "action_horizon_steps": null, "size": "16.2B", "gpu": "NVIDIA A800", "vram": "40 GB", "cameras": [ "EGO-L/R", "Wrist" ] }, "open_sourcing": { "status": "open" }, "training": { "sim_image_data": { "robolab_in_finetuning": true, "robolab_in_training": false, "other_sim_in_training": false }, "simulation_use": "RoboLab sim in fine-tuning. 4 tasks (ThrowAwayAppleTask, RubiksCubesInBinTask, ElectronicsInBinTask, FoodPacking1CansTask) for GRPO fine-tuning." }, "provenance": { "display_date": "Submitted Sep 15, 2026", "robolab_build": "7d45d749 (v0.1.1)", "isaac_versions": "2.2.0 / 5.0.0" }, "results": { "default": { "episodes": 1200, "successes": 475, "sr": 39.6, "score": 55.5, "score_ci_bootstrap": [ 49.2, 61.9 ], "ee_speed_cm_s": 7.69, "ee_speed_sd_cm_s": 1.97, "ee_sparc": -4.77, "by_difficulty": { "simple": { "episodes": 640, "successes": 284, "sr": 44.4, "score": 54.6, "score_ci_bootstrap": [ 45.2, 64.0 ], "sr_ci_cp": [ 40.48137726724518, 48.32073592352406 ] }, "moderate": { "episodes": 390, "successes": 158, "sr": 40.5, "score": 58.1, "score_ci_bootstrap": [ 47.2, 68.8 ], "sr_ci_cp": [ 35.60072625302086, 45.5700306910844 ] }, "complex": { "episodes": 170, "successes": 33, "sr": 19.4, "score": 53.1, "score_ci_bootstrap": [ 40.8, 65.8 ], "sr_ci_cp": [ 13.754956495970369, 26.166826538970376 ] } }, "by_axis": { "procedural": { "episodes": 220, "successes": 57, "sr": 25.9, "score": 46.1, "score_ci_bootstrap": [ 35.6, 56.6 ], "sr_ci_cp": [ 20.25271594238445, 32.22716775375001 ] }, "relational": { "episodes": 420, "successes": 189, "sr": 45.0, "score": 57.9, "score_ci_bootstrap": [ 45.9, 69.6 ], "sr_ci_cp": [ 40.17298944346469, 49.89793263982616 ] }, "visual": { "episodes": 840, "successes": 286, "sr": 34.0, "score": 49.9, "score_ci_bootstrap": [ 42.0, 57.7 ], "sr_ci_cp": [ 30.844292780603766, 37.36336425731106 ] } }, "episodes_per_task": "10", "sr_ci_cp": [ 36.80318867767137, 42.41466057897826 ] }, "vague": { "episodes": 1200, "successes": 379, "sr": 31.6, "score": 48.6, "score_ci_bootstrap": [ 42.2, 54.9 ], "ee_speed_cm_s": 7.45, "ee_speed_sd_cm_s": 1.79, "ee_sparc": -4.76, "by_difficulty": { "simple": { "episodes": 640, "successes": 238, "sr": 37.2, "score": 47.4, "score_ci_bootstrap": [ 38.4, 56.4 ], "sr_ci_cp": [ 33.43155032703355, 41.06223679172491 ] }, "moderate": { "episodes": 390, "successes": 118, "sr": 30.3, "score": 51.2, "score_ci_bootstrap": [ 39.8, 62.7 ], "sr_ci_cp": [ 25.734143539650177, 35.08134988336805 ] }, "complex": { "episodes": 170, "successes": 23, "sr": 13.5, "score": 46.9, "score_ci_bootstrap": [ 34.4, 60.0 ], "sr_ci_cp": [ 8.773695070855647, 19.606196156208735 ] } }, "by_axis": { "procedural": { "episodes": 220, "successes": 35, "sr": 15.9, "score": 38.2, "score_ci_bootstrap": [ 26.2, 50.7 ], "sr_ci_cp": [ 11.33835239397944, 21.424811457750668 ] }, "relational": { "episodes": 420, "successes": 162, "sr": 38.6, "score": 54.2, "score_ci_bootstrap": [ 42.7, 65.9 ], "sr_ci_cp": [ 33.8920700433765, 43.41300580285802 ] }, "visual": { "episodes": 840, "successes": 221, "sr": 26.3, "score": 42.6, "score_ci_bootstrap": [ 35.1, 50.1 ], "sr_ci_cp": [ 23.359822653590406, 29.42647424852475 ] } }, "episodes_per_task": "10", "sr_ci_cp": [ 28.959415355882275, 34.29783264024359 ] } }, "badges": [ "Organizer cross-check" ], "links": [ { "label": "Code", "url": "https://huggingface.co/datasets/MochunniaN1/Atomic-WAM_Robolab/tree/main/Atomic-WAM_RoboLab_repro", "kind": "code" }, { "label": "Checkpoint", "url": "https://huggingface.co/datasets/MochunniaN1/Atomic-WAM_Robolab/tree/main/Atomic-WAM_RoboLab_repro/cosmos-framework/checkpoints", "kind": "model" } ], "table": { "observations": [ "RGB", "P" ] } }, { "id": "oasis-wam", "display_name": "OASIS WAM", "organization": "USTC & Synkrotron", "organization_links": [ { "label": "USTC", "url": "https://en.ustc.edu.cn/" }, { "label": "Synkrotron", "url": "https://www.synkrotron.ai/" } ], "source": "external-submission", "policy": { "type": "VLM+WAM", "type_other": null, "description": "Asynchronous high-level VLM reasoning over current and predicted future observations to produce the next complete manipulation subgoal.", "robot": "DROID (Franka)", "observation_space": [ "RGB", "Proprioception" ], "action_space": [ "Joint position" ], "action_horizon_steps": "32 steps", "size": "WAM 16B / VLM undisclosed", "vram": "WAM: 40 GB", "cameras": [ "EGO-L/R", "Wrist" ], "observation_history_label": "Current", "gpu": null }, "open_sourcing": { "status": "closed" }, "training": { "sim_image_data": { "robolab_in_finetuning": false, "robolab_in_training": false, "other_sim_in_training": true }, "simulation_use": "Other simulation data" }, "provenance": { "display_date": "Submitted Sep 2, 2026", "robolab_build": "9db0aaf0 (v0.3.1)", "isaac_versions": "2.2.0 / 5.0.0" }, "results": { "default": { "episodes": 1200, "successes": 468, "sr": 39.0, "score": 53.7, "score_ci_bootstrap": [ 46.8, 60.3 ], "ee_speed_cm_s": 6.91, "ee_speed_sd_cm_s": 1.81, "ee_sparc": -5.26, "by_difficulty": { "simple": { "episodes": 640, "successes": 265, "sr": 41.4, "score": 50.7, "score_ci_bootstrap": [ 40.8, 60.6 ], "sr_ci_cp": [ 37.55931251248354, 45.332822488710924 ] }, "moderate": { "episodes": 390, "successes": 148, "sr": 37.9, "score": 55.6, "score_ci_bootstrap": [ 44.8, 66.3 ], "sr_ci_cp": [ 33.11205318941769, 42.969789159331754 ] }, "complex": { "episodes": 170, "successes": 55, "sr": 32.4, "score": 60.5, "score_ci_bootstrap": [ 45.8, 74.4 ], "sr_ci_cp": [ 25.391421423232647, 39.942431685884685 ] } }, "by_axis": { "procedural": { "episodes": 220, "successes": 67, "sr": 30.5, "score": 50.9, "score_ci_bootstrap": [ 39.3, 62.6 ], "sr_ci_cp": [ 24.44689236055367, 36.997843567285074 ] }, "relational": { "episodes": 420, "successes": 196, "sr": 46.7, "score": 59.3, "score_ci_bootstrap": [ 47.8, 70.4 ], "sr_ci_cp": [ 41.814582086729665, 51.566028526850594 ] }, "visual": { "episodes": 840, "successes": 248, "sr": 29.5, "score": 45.3, "score_ci_bootstrap": [ 37.4, 53.4 ], "sr_ci_cp": [ 26.455946854296737, 32.736102826518795 ] } }, "episodes_per_task": "10", "sr_ci_cp": [ 36.22851104804708, 41.82554049337039 ] } }, "badges": [ "Organizer cross-check" ], "links": [], "table": { "observations": [ "RGB", "P" ], "size": "16B + VLM", "vram": "40 GB" } }, { "id": "cosmos3", "display_name": "Cosmos3-Nano-Policy", "organization": "NVIDIA Cosmos", "organization_links": [ { "label": "NVIDIA Cosmos", "url": "https://www.nvidia.com/en-us/ai/cosmos/" } ], "source": "internal", "policy": { "type": "WAM", "type_other": null, "description": "An omnimodal world-action model.", "robot": "DROID (Franka)", "observation_space": [ "RGB", "Proprioception" ], "action_space": null, "action_horizon_steps": "32 steps", "size": "16B", "gpu": "NVIDIA L40", "vram": "40 GB", "cameras": [ "EGO-L/R", "Wrist" ], "observation_history_label": "Current" }, "open_sourcing": null, "training": { "sim_image_data": { "robolab_in_finetuning": false, "robolab_in_training": false, "other_sim_in_training": false } }, "provenance": { "display_date": "Evaluated May 22, 2026", "robolab_build": "88976795 (v0.1.0)", "isaac_versions": "2.2.0 / 5.0.0" }, "results": { "default": { "episodes": 1200, "successes": 441, "sr": 36.8, "score": 51.9, "score_ci_bootstrap": [ 45.2, 58.5 ], "ee_speed_cm_s": 7.13, "ee_speed_sd_cm_s": 1.77, "ee_sparc": -5.99, "by_difficulty": { "simple": { "episodes": 640, "successes": 260, "sr": 40.6, "score": 51.0, "score_ci_bootstrap": [ 41.0, 60.7 ], "sr_ci_cp": [ 36.79269817071461, 44.54418288129895 ] }, "moderate": { "episodes": 390, "successes": 138, "sr": 35.4, "score": 52.3, "score_ci_bootstrap": [ 41.6, 63.4 ], "sr_ci_cp": [ 30.63760029822205, 40.355381639643596 ] }, "complex": { "episodes": 170, "successes": 43, "sr": 25.3, "score": 54.4, "score_ci_bootstrap": [ 37.9, 69.9 ], "sr_ci_cp": [ 18.950087401633148, 32.520610481148104 ] } }, "by_axis": { "procedural": { "episodes": 220, "successes": 60, "sr": 27.3, "score": 46.8, "score_ci_bootstrap": [ 35.6, 58.0 ], "sr_ci_cp": [ 21.50305898341156, 33.66616829476825 ] }, "relational": { "episodes": 420, "successes": 207, "sr": 49.3, "score": 60.9, "score_ci_bootstrap": [ 49.3, 71.9 ], "sr_ci_cp": [ 44.404966437614895, 54.176592312471136 ] }, "visual": { "episodes": 840, "successes": 219, "sr": 26.1, "score": 42.9, "score_ci_bootstrap": [ 35.0, 50.8 ], "sr_ci_cp": [ 23.13126551229836, 29.180534313486355 ] } }, "episodes_per_task": "10", "sr_ci_cp": [ 34.01580882235194, 39.54931336596538 ] }, "vague": { "episodes": 1200, "successes": 247, "sr": 20.6, "score": 33.2, "score_ci_bootstrap": [ 26.7, 40.0 ], "ee_speed_cm_s": 6.37, "ee_speed_sd_cm_s": 1.59, "ee_sparc": -5.76, "by_difficulty": { "simple": { "episodes": 640, "successes": 149, "sr": 23.3, "score": 30.1, "score_ci_bootstrap": [ 21.0, 39.6 ], "sr_ci_cp": [ 20.058412028546893, 26.75273374225995 ] }, "moderate": { "episodes": 390, "successes": 91, "sr": 23.3, "score": 43.1, "score_ci_bootstrap": [ 31.2, 55.2 ], "sr_ci_cp": [ 19.223667966491046, 27.8530209171081 ] }, "complex": { "episodes": 170, "successes": 7, "sr": 4.1, "score": 22.5, "score_ci_bootstrap": [ 11.3, 34.3 ], "sr_ci_cp": [ 1.671287148479946, 8.299653645403966 ] } }, "by_axis": { "procedural": { "episodes": 220, "successes": 30, "sr": 13.6, "score": 36.5, "score_ci_bootstrap": [ 24.8, 48.3 ], "sr_ci_cp": [ 9.393166190448925, 18.89116474031298 ] }, "relational": { "episodes": 420, "successes": 136, "sr": 32.4, "score": 46.4, "score_ci_bootstrap": [ 33.9, 59.0 ], "sr_ci_cp": [ 27.924722790676682, 37.087588363396755 ] }, "visual": { "episodes": 840, "successes": 113, "sr": 13.5, "score": 26.1, "score_ci_bootstrap": [ 19.2, 33.4 ], "sr_ci_cp": [ 11.216827186540094, 15.947833010616733 ] } }, "episodes_per_task": "10", "sr_ci_cp": [ 18.328426961217065, 22.983360731076505 ] }, "specific": { "episodes": 1200, "successes": 476, "sr": 39.7, "score": 54.8, "score_ci_bootstrap": [ 48.3, 61.2 ], "ee_speed_cm_s": 7.31, "ee_speed_sd_cm_s": 1.94, "ee_sparc": -5.85, "by_difficulty": { "simple": { "episodes": 640, "successes": 269, "sr": 42.0, "score": 52.3, "score_ci_bootstrap": [ 42.7, 61.9 ], "sr_ci_cp": [ 38.17331440349369, 45.96302491421888 ] }, "moderate": { "episodes": 390, "successes": 157, "sr": 40.3, "score": 57.5, "score_ci_bootstrap": [ 46.2, 68.5 ], "sr_ci_cp": [ 35.351234172087324, 45.31062928694226 ] }, "complex": { "episodes": 170, "successes": 50, "sr": 29.4, "score": 58.2, "score_ci_bootstrap": [ 43.3, 72.8 ], "sr_ci_cp": [ 22.68332365690038, 36.87380565745458 ] } }, "by_axis": { "procedural": { "episodes": 220, "successes": 63, "sr": 28.6, "score": 47.0, "score_ci_bootstrap": [ 34.6, 59.2 ], "sr_ci_cp": [ 22.76031616195803, 35.0983564156398 ] }, "relational": { "episodes": 420, "successes": 216, "sr": 51.4, "score": 62.7, "score_ci_bootstrap": [ 51.3, 73.8 ], "sr_ci_cp": [ 46.53408370117074, 56.30279860653409 ] }, "visual": { "episodes": 840, "successes": 244, "sr": 29.0, "score": 46.5, "score_ci_bootstrap": [ 38.8, 54.5 ], "sr_ci_cp": [ 25.996056720969936, 32.246988055757484 ] } }, "episodes_per_task": "10", "sr_ci_cp": [ 36.885319098095465, 42.49878701158055 ] } }, "badges": [ "Organizer cross-check" ], "links": [ { "label": "Paper", "url": "https://research.nvidia.com/labs/cosmos-lab/cosmos3/technical-report.pdf", "kind": "paper" }, { "label": "Project page", "url": "https://huggingface.co/nvidia/Cosmos3-Nano-Policy-DROID", "kind": "project" } ], "table": { "observations": [ "RGB", "P" ] } }, { "id": "phoenix", "display_name": "Phoenix", "organization": "Geodesic", "organization_links": [ { "label": "Geodesic", "url": "https://www.geodesicos.com/" } ], "source": "external-submission", "policy": { "type": "other", "type_other": "TAMP+FM", "description": null, "robot": "DROID (Franka + Robotiq 2F-85)", "observation_space": [ "RGB", "Depth", "Proprioception" ], "action_space": [ "Joint position" ], "action_horizon_steps": "N/A", "size": "~250M", "vram": "14 GB", "cameras": [ "EGO-L", "Wrist" ], "observation_history_label": "Current", "gpu": null }, "open_sourcing": { "status": "closed" }, "training": { "sim_image_data": { "robolab_in_finetuning": false, "robolab_in_training": false, "other_sim_in_training": true }, "simulation_use": "Other simulation data" }, "provenance": { "display_date": "Submitted Aug 2, 2026", "robolab_build": "0aef241f (v0.2.1)", "isaac_versions": "2.2.0 / 5.0.0" }, "results": { "default": { "episodes": 1200, "successes": 413, "sr": 34.4, "score": 45.9, "score_ci_bootstrap": [ 39.2, 52.6 ], "ee_speed_cm_s": 6.13, "ee_speed_sd_cm_s": 4.56, "ee_sparc": -5.55, "by_difficulty": { "simple": { "episodes": 640, "successes": 274, "sr": 42.8, "score": 51.9, "score_ci_bootstrap": [ 42.6, 61.1 ], "sr_ci_cp": [ 38.9416981457499, 46.74989787589081 ] }, "moderate": { "episodes": 390, "successes": 100, "sr": 25.6, "score": 40.0, "score_ci_bootstrap": [ 28.7, 51.8 ], "sr_ci_cp": [ 21.378397477424187, 30.277753814305363 ] }, "complex": { "episodes": 170, "successes": 39, "sr": 22.9, "score": 37.0, "score_ci_bootstrap": [ 20.5, 54.4 ], "sr_ci_cp": [ 16.850712099288558, 29.99987091680668 ] } }, "by_axis": { "procedural": { "episodes": 220, "successes": 48, "sr": 21.8, "score": 30.3, "score_ci_bootstrap": [ 17.2, 44.4 ], "sr_ci_cp": [ 16.54695694654881, 27.865672247360994 ] }, "relational": { "episodes": 420, "successes": 105, "sr": 25.0, "score": 37.3, "score_ci_bootstrap": [ 26.2, 49.0 ], "sr_ci_cp": [ 20.928505899810325, 29.427742049048494 ] }, "visual": { "episodes": 840, "successes": 282, "sr": 33.6, "score": 45.4, "score_ci_bootstrap": [ 37.3, 53.5 ], "sr_ci_cp": [ 30.38077247641716, 36.877868789824014 ] } }, "episodes_per_task": "10", "sr_ci_cp": [ 31.727902092091547, 37.18204358094134 ] } }, "badges": [], "links": [], "table": { "observations": [ "RGB", "D", "P" ] } }, { "id": "bimind", "display_name": "BiMind v0.1", "organization": "Bicameral Robotics", "organization_links": [ { "label": "Bicameral Robotics", "url": "https://www.bicameralrobotics.ai/" } ], "source": "external-submission", "policy": { "type": "VLA", "type_other": null, "description": "A VLA with built-in spatial reasoning architecture", "robot": "DROID (Franka)", "observation_space": [ "RGB", "Proprioception" ], "action_space": [ "Joint position" ], "action_horizon_steps": "N/A", "size": "7B", "vram": "14 GiB", "cameras": [ "EGO-L", "Wrist" ], "observation_history_label": "Current", "gpu": null }, "open_sourcing": { "status": "closed" }, "training": { "sim_image_data": { "robolab_in_finetuning": true, "robolab_in_training": true, "other_sim_in_training": true }, "simulation_use": "RoboLab/IsaacLab/IsaacSim data in training and fine-tuning; Other simulation data. Simulation comprises approximately 10% of all samples across pretraining and fine-tuning." }, "provenance": { "display_date": "Submitted Aug 19, 2026", "robolab_build": "9db0aaf0 (v0.3.1)", "isaac_versions": "2.2.0 / 5.0.0" }, "results": { "default": { "episodes": 1200, "successes": 400, "sr": 33.3, "score": 47.2, "score_ci_bootstrap": [ 41.0, 53.5 ], "ee_speed_cm_s": 5.26, "ee_speed_sd_cm_s": 1.72, "ee_sparc": -5.88, "by_difficulty": { "simple": { "episodes": 640, "successes": 248, "sr": 38.8, "score": 46.2, "score_ci_bootstrap": [ 37.4, 55.0 ], "sr_ci_cp": [ 34.95690109612401, 42.6473780244827 ] }, "moderate": { "episodes": 390, "successes": 128, "sr": 32.8, "score": 50.9, "score_ci_bootstrap": [ 40.1, 61.4 ], "sr_ci_cp": [ 28.178017002076906, 37.7261752175281 ] }, "complex": { "episodes": 170, "successes": 24, "sr": 14.1, "score": 42.5, "score_ci_bootstrap": [ 28.8, 56.5 ], "sr_ci_cp": [ 9.259575892275548, 20.273992519429008 ] } }, "by_axis": { "procedural": { "episodes": 220, "successes": 47, "sr": 21.4, "score": 33.3, "score_ci_bootstrap": [ 20.8, 46.6 ], "sr_ci_cp": [ 16.13976678668114, 27.37658959641287 ] }, "relational": { "episodes": 420, "successes": 150, "sr": 35.7, "score": 51.7, "score_ci_bootstrap": [ 41.2, 62.0 ], "sr_ci_cp": [ 31.127557332751437, 40.503889596407724 ] }, "visual": { "episodes": 840, "successes": 247, "sr": 29.4, "score": 43.0, "score_ci_bootstrap": [ 35.6, 50.5 ], "sr_ci_cp": [ 26.340936295223006, 32.61386202759613 ] } }, "episodes_per_task": "10", "sr_ci_cp": [ 30.668077908667467, 36.080539777992676 ] }, "vague": { "episodes": 1200, "successes": 363, "sr": 30.2, "score": 44.9, "score_ci_bootstrap": [ 38.7, 51.2 ], "ee_speed_cm_s": 5.24, "ee_speed_sd_cm_s": 1.72, "ee_sparc": -5.92, "by_difficulty": { "simple": { "episodes": 640, "successes": 218, "sr": 34.1, "score": 42.0, "score_ci_bootstrap": [ 33.1, 50.9 ], "sr_ci_cp": [ 30.393500316516047, 37.879336026725866 ] }, "moderate": { "episodes": 390, "successes": 111, "sr": 28.5, "score": 47.4, "score_ci_bootstrap": [ 37.2, 57.9 ], "sr_ci_cp": [ 24.033345666996457, 33.22012611616894 ] }, "complex": { "episodes": 170, "successes": 34, "sr": 20.0, "score": 50.5, "score_ci_bootstrap": [ 35.9, 65.4 ], "sr_ci_cp": [ 14.26607006948038, 26.81037347214086 ] } }, "by_axis": { "procedural": { "episodes": 220, "successes": 35, "sr": 15.9, "score": 28.8, "score_ci_bootstrap": [ 18.8, 39.2 ], "sr_ci_cp": [ 11.33835239397944, 21.424811457750668 ] }, "relational": { "episodes": 420, "successes": 154, "sr": 36.7, "score": 53.0, "score_ci_bootstrap": [ 41.8, 63.8 ], "sr_ci_cp": [ 32.04714192192477, 41.47550724148241 ] }, "visual": { "episodes": 840, "successes": 206, "sr": 24.5, "score": 39.8, "score_ci_bootstrap": [ 32.7, 47.2 ], "sr_ci_cp": [ 21.648447507400302, 27.579133770766344 ] } }, "episodes_per_task": "10", "sr_ci_cp": [ 27.660459309606033, 32.93670345885545 ] }, "specific": { "episodes": 1200, "successes": 367, "sr": 30.6, "score": 44.7, "score_ci_bootstrap": [ 38.5, 50.9 ], "ee_speed_cm_s": 5.05, "ee_speed_sd_cm_s": 1.71, "ee_sparc": -6.14, "by_difficulty": { "simple": { "episodes": 640, "successes": 229, "sr": 35.8, "score": 43.0, "score_ci_bootstrap": [ 33.9, 52.4 ], "sr_ci_cp": [ 32.0623037041888, 39.632049512937215 ] }, "moderate": { "episodes": 390, "successes": 121, "sr": 31.0, "score": 48.4, "score_ci_bootstrap": [ 37.7, 59.1 ], "sr_ci_cp": [ 26.465597090863124, 35.87649502719124 ] }, "complex": { "episodes": 170, "successes": 17, "sr": 10.0, "score": 42.3, "score_ci_bootstrap": [ 29.6, 55.6 ], "sr_ci_cp": [ 5.934026029560021, 15.527868088096037 ] } }, "by_axis": { "procedural": { "episodes": 220, "successes": 47, "sr": 21.4, "score": 32.6, "score_ci_bootstrap": [ 20.7, 45.2 ], "sr_ci_cp": [ 16.13976678668114, 27.37658959641287 ] }, "relational": { "episodes": 420, "successes": 130, "sr": 31.0, "score": 48.6, "score_ci_bootstrap": [ 37.8, 59.5 ], "sr_ci_cp": [ 26.5598768626792, 35.61570317657847 ] }, "visual": { "episodes": 840, "successes": 215, "sr": 25.6, "score": 39.8, "score_ci_bootstrap": [ 32.8, 47.2 ], "sr_ci_cp": [ 22.674491525296858, 28.688315622776873 ] } }, "episodes_per_task": "10", "sr_ci_cp": [ 27.984959049470838, 33.27722449820267 ] } }, "badges": [], "links": [], "table": { "observations": [ "RGB", "P" ] } }, { "id": "volo-pi05", "display_name": "VoLo", "organization": "NVIDIA LPR", "organization_links": [ { "label": "NVIDIA LPR", "url": "https://research.nvidia.com/labs/lpr/" } ], "source": "external-submission", "policy": { "type": "other", "type_other": "Agent", "description": "A physical-orchestration agent that uses a VLM to plan, monitor, and recover while directing a VLA or WAM as an interruptible manipulation tool.", "robot": "DROID (Franka)", "observation_space": [ "RGB", "Depth", "Proprioception" ], "action_space": [ "Joint position" ], "action_horizon_steps": null, "size": "undisclosed", "gpu": null, "vram": "80 GB", "cameras": [ "EGO-L", "Wrist" ], "observation_history_label": null }, "open_sourcing": { "status": "open" }, "training": { "sim_image_data": { "robolab_in_finetuning": false, "robolab_in_training": false, "other_sim_in_training": true }, "simulation_use": "Other simulation data" }, "provenance": { "display_date": "Submitted Sep 8, 2026", "robolab_build": "f72d3944 (v0.3.0)", "isaac_versions": "Not reported" }, "results": { "default": { "episodes": 1200, "successes": 339, "sr": 28.2, "score": 44.1, "score_ci_bootstrap": [ 37.7, 50.6 ], "ee_speed_cm_s": 5.5, "ee_speed_sd_cm_s": 1.67, "ee_sparc": -8.92, "by_difficulty": { "simple": { "episodes": 640, "successes": 209, "sr": 32.7, "score": 43.7, "score_ci_bootstrap": [ 34.3, 53.2 ], "sr_ci_cp": [ 29.032096261798245, 36.44132996995873 ] }, "moderate": { "episodes": 390, "successes": 100, "sr": 25.6, "score": 46.7, "score_ci_bootstrap": [ 36.7, 56.6 ], "sr_ci_cp": [ 21.378397477424187, 30.277753814305363 ] }, "complex": { "episodes": 170, "successes": 30, "sr": 17.6, "score": 39.8, "score_ci_bootstrap": [ 25.6, 54.0 ], "sr_ci_cp": [ 12.234466043283678, 24.22375162326456 ] } }, "by_axis": { "procedural": { "episodes": 220, "successes": 36, "sr": 16.4, "score": 40.4, "score_ci_bootstrap": [ 28.9, 51.7 ], "sr_ci_cp": [ 11.73177644415504, 21.927300647000084 ] }, "relational": { "episodes": 420, "successes": 134, "sr": 31.9, "score": 44.1, "score_ci_bootstrap": [ 32.5, 55.8 ], "sr_ci_cp": [ 27.46923858966759, 36.597492421386406 ] }, "visual": { "episodes": 840, "successes": 196, "sr": 23.3, "score": 39.3, "score_ci_bootstrap": [ 31.9, 46.5 ], "sr_ci_cp": [ 20.511259010584112, 26.343862787035327 ] } }, "episodes_per_task": "10", "sr_ci_cp": [ 25.716917989100175, 30.890128342163507 ] }, "vague": { "episodes": 1200, "successes": 354, "sr": 29.5, "score": 45.4, "score_ci_bootstrap": [ 38.9, 51.8 ], "ee_speed_cm_s": 5.51, "ee_speed_sd_cm_s": 1.62, "ee_sparc": -9.03, "by_difficulty": { "simple": { "episodes": 640, "successes": 222, "sr": 34.7, "score": 45.3, "score_ci_bootstrap": [ 36.0, 54.5 ], "sr_ci_cp": [ 30.999727495718222, 38.51729494333508 ] }, "moderate": { "episodes": 390, "successes": 96, "sr": 24.6, "score": 46.4, "score_ci_bootstrap": [ 35.4, 57.3 ], "sr_ci_cp": [ 20.418697276626492, 29.20211672186253 ] }, "complex": { "episodes": 170, "successes": 36, "sr": 21.2, "score": 43.2, "score_ci_bootstrap": [ 28.6, 57.7 ], "sr_ci_cp": [ 15.29429348063218, 28.091650128230338 ] } }, "by_axis": { "procedural": { "episodes": 220, "successes": 40, "sr": 18.2, "score": 42.2, "score_ci_bootstrap": [ 29.9, 54.4 ], "sr_ci_cp": [ 13.318490196087032, 23.924627047600183 ] }, "relational": { "episodes": 420, "successes": 131, "sr": 31.2, "score": 44.3, "score_ci_bootstrap": [ 32.4, 56.2 ], "sr_ci_cp": [ 26.78701440879861, 35.861352164016196 ] }, "visual": { "episodes": 840, "successes": 205, "sr": 24.4, "score": 40.3, "score_ci_bootstrap": [ 32.9, 47.7 ], "sr_ci_cp": [ 21.534590928780233, 27.45574372420665 ] } }, "episodes_per_task": "10", "sr_ci_cp": [ 26.930929436614466, 32.16993795433842 ] }, "specific": { "episodes": 1200, "successes": 377, "sr": 31.4, "score": 46.1, "score_ci_bootstrap": [ 39.6, 52.6 ], "ee_speed_cm_s": 5.49, "ee_speed_sd_cm_s": 1.67, "ee_sparc": -9.04, "by_difficulty": { "simple": { "episodes": 640, "successes": 231, "sr": 36.1, "score": 46.9, "score_ci_bootstrap": [ 37.1, 56.6 ], "sr_ci_cp": [ 32.366284136962314, 39.950164704703546 ] }, "moderate": { "episodes": 390, "successes": 108, "sr": 27.7, "score": 46.4, "score_ci_bootstrap": [ 35.6, 57.1 ], "sr_ci_cp": [ 23.307058632438547, 32.41985250041719 ] }, "complex": { "episodes": 170, "successes": 38, "sr": 22.4, "score": 42.3, "score_ci_bootstrap": [ 27.9, 56.8 ], "sr_ci_cp": [ 16.330101784083688, 29.36555413522074 ] } }, "by_axis": { "procedural": { "episodes": 220, "successes": 50, "sr": 22.7, "score": 41.7, "score_ci_bootstrap": [ 29.0, 54.7 ], "sr_ci_cp": [ 17.36420849093523, 28.84102596880829 ] }, "relational": { "episodes": 420, "successes": 138, "sr": 32.9, "score": 45.5, "score_ci_bootstrap": [ 34.2, 57.2 ], "sr_ci_cp": [ 28.38073414249842, 37.577160074346885 ] }, "visual": { "episodes": 840, "successes": 227, "sr": 27.0, "score": 41.8, "score_ci_bootstrap": [ 33.9, 49.8 ], "sr_ci_cp": [ 24.046162829380172, 30.163628110840214 ] } }, "episodes_per_task": "10", "sr_ci_cp": [ 28.796907308737847, 34.12782973425447 ] } }, "badges": [], "links": [ { "label": "Paper", "url": "https://arxiv.org/abs/2606.07723", "kind": "paper" }, { "label": "Code", "url": "https://github.com/NVlabs/VoLoAgent", "kind": "code" }, { "label": "Project page", "url": "https://chicychen.github.io/VoLo/", "kind": "project" } ], "table": { "observations": [ "RGB", "D", "P" ], "size": "—" } }, { "id": "pi05", "display_name": "π0.5", "organization": "Physical Intelligence", "organization_links": [ { "label": "Physical Intelligence", "url": "https://www.pi.website/" } ], "source": "internal", "policy": { "type": "VLA", "type_other": null, "description": "A VLA model trained on diverse robot, semantic, and web data for long-horizon manipulation in previously unseen environments.", "robot": "DROID (Franka)", "observation_space": [ "RGB", "Proprioception" ], "action_space": [ "Joint position" ], "action_horizon_steps": "15 steps", "size": "3.3B", "gpu": "NVIDIA L40S", "vram": ">8 GB (OpenPI inference requirement)", "cameras": [ "EGO-L", "Wrist" ], "observation_history_label": "Current" }, "open_sourcing": null, "training": { "sim_image_data": { "robolab_in_finetuning": false, "robolab_in_training": false, "other_sim_in_training": false } }, "provenance": { "display_date": "Evaluated May 9, 2026", "robolab_build": "09545754 (v0.1.0)", "isaac_versions": "2.2.0 / 5.0.0" }, "results": { "default": { "episodes": 1200, "successes": 336, "sr": 28.0, "score": 43.4, "score_ci_bootstrap": [ 37.3, 49.5 ], "ee_speed_cm_s": 5.35, "ee_speed_sd_cm_s": 1.59, "ee_sparc": -8.34, "by_difficulty": { "simple": { "episodes": 640, "successes": 190, "sr": 29.7, "score": 38.9, "score_ci_bootstrap": [ 30.6, 47.6 ], "sr_ci_cp": [ 26.170352982253696, 33.39325358205749 ] }, "moderate": { "episodes": 390, "successes": 123, "sr": 31.5, "score": 50.5, "score_ci_bootstrap": [ 40.1, 61.3 ], "sr_ci_cp": [ 26.954055394960047, 36.40577461147368 ] }, "complex": { "episodes": 170, "successes": 23, "sr": 13.5, "score": 44.1, "score_ci_bootstrap": [ 29.6, 58.3 ], "sr_ci_cp": [ 8.773695070855647, 19.606196156208735 ] } }, "by_axis": { "procedural": { "episodes": 220, "successes": 48, "sr": 21.8, "score": 41.6, "score_ci_bootstrap": [ 29.3, 54.0 ], "sr_ci_cp": [ 16.54695694654881, 27.865672247360994 ] }, "relational": { "episodes": 420, "successes": 142, "sr": 33.8, "score": 46.8, "score_ci_bootstrap": [ 35.7, 58.2 ], "sr_ci_cp": [ 29.294314761797015, 38.55475382055055 ] }, "visual": { "episodes": 840, "successes": 200, "sr": 23.8, "score": 38.8, "score_ci_bootstrap": [ 31.6, 46.2 ], "sr_ci_cp": [ 20.965764004398963, 26.83833974042182 ] } }, "episodes_per_task": "10", "sr_ci_cp": [ 25.474402972593552, 30.6338798843704 ] }, "vague": { "episodes": 1200, "successes": 183, "sr": 15.2, "score": 28.1, "score_ci_bootstrap": [ 22.1, 34.2 ], "ee_speed_cm_s": 5.35, "ee_speed_sd_cm_s": 1.41, "ee_sparc": -8.54, "by_difficulty": { "simple": { "episodes": 640, "successes": 104, "sr": 16.2, "score": 24.4, "score_ci_bootstrap": [ 16.4, 33.2 ], "sr_ci_cp": [ 13.474830638335344, 19.340605717592858 ] }, "moderate": { "episodes": 390, "successes": 70, "sr": 17.9, "score": 37.6, "score_ci_bootstrap": [ 27.0, 48.7 ], "sr_ci_cp": [ 14.267526622965665, 22.124600924682756 ] }, "complex": { "episodes": 170, "successes": 9, "sr": 5.3, "score": 20.3, "score_ci_bootstrap": [ 9.7, 32.8 ], "sr_ci_cp": [ 2.4490751990225483, 9.811761320777793 ] } }, "by_axis": { "procedural": { "episodes": 220, "successes": 24, "sr": 10.9, "score": 30.6, "score_ci_bootstrap": [ 20.4, 41.4 ], "sr_ci_cp": [ 7.1158873131693845, 15.796074132057777 ] }, "relational": { "episodes": 420, "successes": 101, "sr": 24.0, "score": 38.8, "score_ci_bootstrap": [ 27.1, 51.0 ], "sr_ci_cp": [ 20.03658641447482, 28.428643513964268 ] }, "visual": { "episodes": 840, "successes": 103, "sr": 12.3, "score": 25.0, "score_ci_bootstrap": [ 18.0, 32.3 ], "sr_ci_cp": [ 10.11977133698318, 14.672748056034877 ] } }, "episodes_per_task": "10", "sr_ci_cp": [ 13.261367769149802, 17.41055637646323 ] }, "specific": { "episodes": 1200, "successes": 337, "sr": 28.1, "score": 42.5, "score_ci_bootstrap": [ 36.7, 48.6 ], "ee_speed_cm_s": 5.32, "ee_speed_sd_cm_s": 1.59, "ee_sparc": -8.57, "by_difficulty": { "simple": { "episodes": 640, "successes": 191, "sr": 29.8, "score": 37.6, "score_ci_bootstrap": [ 29.3, 46.1 ], "sr_ci_cp": [ 26.320538478744815, 33.554109441388704 ] }, "moderate": { "episodes": 390, "successes": 121, "sr": 31.0, "score": 49.0, "score_ci_bootstrap": [ 38.2, 59.7 ], "sr_ci_cp": [ 26.465597090863124, 35.87649502719124 ] }, "complex": { "episodes": 170, "successes": 25, "sr": 14.7, "score": 46.5, "score_ci_bootstrap": [ 32.4, 60.0 ], "sr_ci_cp": [ 9.74852841315704, 20.9388600948618 ] } }, "by_axis": { "procedural": { "episodes": 220, "successes": 45, "sr": 20.5, "score": 37.0, "score_ci_bootstrap": [ 24.4, 50.0 ], "sr_ci_cp": [ 15.328370029378815, 26.395505912076082 ] }, "relational": { "episodes": 420, "successes": 131, "sr": 31.2, "score": 44.5, "score_ci_bootstrap": [ 33.6, 55.6 ], "sr_ci_cp": [ 26.78701440879861, 35.861352164016196 ] }, "visual": { "episodes": 840, "successes": 210, "sr": 25.0, "score": 39.2, "score_ci_bootstrap": [ 32.1, 46.3 ], "sr_ci_cp": [ 22.10417282007604, 28.07239637536903 ] } }, "episodes_per_task": "10", "sr_ci_cp": [ 25.555230511308846, 30.71930680847596 ] } }, "badges": [ "Organizer cross-check" ], "links": [ { "label": "Project page", "url": "https://www.physicalintelligence.company/blog/pi05", "kind": "project" } ], "table": { "observations": [ "RGB", "P" ], "vram": ">8 GB" } }, { "id": "dreamzero", "display_name": "DreamZero", "organization": "NVIDIA GEAR", "organization_links": [ { "label": "NVIDIA GEAR", "url": "https://research.nvidia.com/labs/gear/" } ], "source": "internal", "policy": { "type": "WAM", "type_other": null, "description": "A video-diffusion world-action model that jointly predicts future visual states and robot actions for real-time, zero-shot manipulation.", "robot": "DROID (Franka)", "observation_space": [ "RGB", "Proprioception" ], "action_space": null, "action_horizon_steps": "24 steps", "size": "14B", "gpu": "NVIDIA L40", "vram": "2x80 GB (H100 public reference setup)", "cameras": [ "EGO-L/R", "Wrist" ], "observation_history_label": "Current" }, "open_sourcing": null, "training": { "sim_image_data": { "robolab_in_finetuning": false, "robolab_in_training": false, "other_sim_in_training": false } }, "provenance": { "display_date": "Evaluated May 16, 2026", "robolab_build": "f7701fd5 (v0.1.0)", "isaac_versions": "2.2.0 / 5.0.0" }, "results": { "default": { "episodes": 1200, "successes": 308, "sr": 25.7, "score": 39.8, "score_ci_bootstrap": [ 33.1, 46.5 ], "ee_speed_cm_s": 3.64, "ee_speed_sd_cm_s": 1.41, "ee_sparc": -6.41, "by_difficulty": { "simple": { "episodes": 640, "successes": 167, "sr": 26.1, "score": 35.5, "score_ci_bootstrap": [ 26.4, 45.1 ], "sr_ci_cp": [ 22.730241617845927, 29.679481638520954 ] }, "moderate": { "episodes": 390, "successes": 117, "sr": 30.0, "score": 48.4, "score_ci_bootstrap": [ 37.1, 59.7 ], "sr_ci_cp": [ 25.49065937332436, 34.81597009792536 ] }, "complex": { "episodes": 170, "successes": 24, "sr": 14.1, "score": 35.9, "score_ci_bootstrap": [ 21.5, 51.1 ], "sr_ci_cp": [ 9.259575892275548, 20.273992519429008 ] } }, "by_axis": { "procedural": { "episodes": 220, "successes": 58, "sr": 26.4, "score": 42.3, "score_ci_bootstrap": [ 30.0, 55.3 ], "sr_ci_cp": [ 20.66870809177282, 32.707611282971015 ] }, "relational": { "episodes": 420, "successes": 140, "sr": 33.3, "score": 45.5, "score_ci_bootstrap": [ 33.7, 57.1 ], "sr_ci_cp": [ 28.837266693777526, 38.066213374380084 ] }, "visual": { "episodes": 840, "successes": 147, "sr": 17.5, "score": 31.9, "score_ci_bootstrap": [ 24.6, 39.8 ], "sr_ci_cp": [ 14.988577194347242, 20.241781534827524 ] } }, "episodes_per_task": "10", "sr_ci_cp": [ 23.215747551188834, 28.23742303404449 ] }, "vague": { "episodes": 1200, "successes": 179, "sr": 14.9, "score": 27.0, "score_ci_bootstrap": [ 21.0, 33.1 ], "ee_speed_cm_s": 3.11, "ee_speed_sd_cm_s": 1.24, "ee_sparc": -6.57, "by_difficulty": { "simple": { "episodes": 640, "successes": 96, "sr": 15.0, "score": 21.4, "score_ci_bootstrap": [ 14.0, 29.6 ], "sr_ci_cp": [ 12.322123592457178, 18.005361553445645 ] }, "moderate": { "episodes": 390, "successes": 76, "sr": 19.5, "score": 39.7, "score_ci_bootstrap": [ 28.1, 51.7 ], "sr_ci_cp": [ 15.67223729079214, 23.77246845370142 ] }, "complex": { "episodes": 170, "successes": 7, "sr": 4.1, "score": 19.1, "score_ci_bootstrap": [ 9.7, 30.3 ], "sr_ci_cp": [ 1.671287148479946, 8.299653645403966 ] } }, "by_axis": { "procedural": { "episodes": 220, "successes": 35, "sr": 15.9, "score": 37.1, "score_ci_bootstrap": [ 25.8, 49.0 ], "sr_ci_cp": [ 11.33835239397944, 21.424811457750668 ] }, "relational": { "episodes": 420, "successes": 80, "sr": 19.0, "score": 32.3, "score_ci_bootstrap": [ 20.8, 43.9 ], "sr_ci_cp": [ 15.40252158224187, 23.135383801197282 ] }, "visual": { "episodes": 840, "successes": 80, "sr": 9.5, "score": 22.0, "score_ci_bootstrap": [ 15.7, 28.9 ], "sr_ci_cp": [ 7.624013546760544, 11.712932234967099 ] } }, "episodes_per_task": "10", "sr_ci_cp": [ 12.947232251286847, 17.059715246437133 ] }, "specific": { "episodes": 1200, "successes": 287, "sr": 23.9, "score": 40.0, "score_ci_bootstrap": [ 34.0, 46.3 ], "ee_speed_cm_s": 3.81, "ee_speed_sd_cm_s": 1.41, "ee_sparc": -6.37, "by_difficulty": { "simple": { "episodes": 640, "successes": 164, "sr": 25.6, "score": 36.6, "score_ci_bootstrap": [ 28.0, 45.7 ], "sr_ci_cp": [ 22.283627655559364, 29.192991428334068 ] }, "moderate": { "episodes": 390, "successes": 104, "sr": 26.7, "score": 45.8, "score_ci_bootstrap": [ 35.0, 57.0 ], "sr_ci_cp": [ 22.341223621775672, 31.350294362370484 ] }, "complex": { "episodes": 170, "successes": 18, "sr": 10.6, "score": 39.6, "score_ci_bootstrap": [ 27.8, 51.6 ], "sr_ci_cp": [ 6.397085437583952, 16.217198997531433 ] } }, "by_axis": { "procedural": { "episodes": 220, "successes": 45, "sr": 20.5, "score": 37.2, "score_ci_bootstrap": [ 27.3, 47.9 ], "sr_ci_cp": [ 15.328370029378815, 26.395505912076082 ] }, "relational": { "episodes": 420, "successes": 122, "sr": 29.0, "score": 40.9, "score_ci_bootstrap": [ 29.6, 52.7 ], "sr_ci_cp": [ 24.747773937790548, 33.645545683175385 ] }, "visual": { "episodes": 840, "successes": 148, "sr": 17.6, "score": 35.2, "score_ci_bootstrap": [ 28.2, 42.3 ], "sr_ci_cp": [ 15.100354134432825, 20.367238321943205 ] } }, "episodes_per_task": "10", "sr_ci_cp": [ 21.52777777778082, 26.43407718721918 ] } }, "badges": [ "Organizer cross-check" ], "links": [ { "label": "Project page", "url": "https://dreamzero0.github.io/", "kind": "project" } ], "table": { "observations": [ "RGB", "P" ], "vram": "2×80 GB" } }, { "id": "cosmos3-edge", "display_name": "Cosmos3-Edge-Policy", "organization": "NVIDIA Cosmos", "organization_links": [ { "label": "NVIDIA Cosmos", "url": "https://www.nvidia.com/en-us/ai/cosmos/" } ], "source": "external-submission", "policy": { "type": "WAM", "type_other": null, "description": "A compact 4B Cosmos 3 world-action policy.", "robot": "DROID (Franka)", "observation_space": [ "RGB", "Proprioception" ], "action_space": null, "action_horizon_steps": "32 steps", "size": "4B", "gpu": "NVIDIA L40", "vram": "9.20 GB (public BF16 deployment peak)", "cameras": [ "EGO-L/R", "Wrist" ], "observation_history_label": "Current" }, "open_sourcing": null, "training": { "sim_image_data": { "robolab_in_finetuning": false, "robolab_in_training": false, "other_sim_in_training": false } }, "provenance": { "display_date": "Evaluated Jul 11, 2026", "robolab_build": "454152dd (v0.2.0)", "isaac_versions": "2.2.0 / 5.0.0" }, "results": { "default": { "episodes": 1200, "successes": 275, "sr": 22.9, "score": null, "score_ci_bootstrap": null, "ee_speed_cm_s": 7.04, "ee_speed_sd_cm_s": 1.93, "ee_sparc": -5.46, "by_difficulty": { "simple": { "episodes": 640, "successes": 164, "sr": 25.6, "score": null, "score_ci_bootstrap": null, "sr_ci_cp": [ 22.283627655559364, 29.192991428334068 ] }, "moderate": { "episodes": 390, "successes": 91, "sr": 23.3, "score": null, "score_ci_bootstrap": null, "sr_ci_cp": [ 19.223667966491046, 27.8530209171081 ] }, "complex": { "episodes": 170, "successes": 20, "sr": 11.8, "score": null, "score_ci_bootstrap": null, "sr_ci_cp": [ 7.336287202190826, 17.583596532238364 ] } }, "by_axis": { "procedural": { "episodes": 220, "successes": 33, "sr": 15.0, "score": null, "score_ci_bootstrap": null, "sr_ci_cp": [ 10.555727839495718, 20.415745418494705 ] }, "relational": { "episodes": 420, "successes": 112, "sr": 26.7, "score": null, "score_ci_bootstrap": null, "sr_ci_cp": [ 22.495716800902418, 31.16986978479798 ] }, "visual": { "episodes": 840, "successes": 141, "sr": 16.8, "score": null, "score_ci_bootstrap": null, "sr_ci_cp": [ 14.318844680767528, 19.48811900136011 ] } }, "episodes_per_task": "10", "sr_ci_cp": [ 20.565702136520358, 25.401123873738978 ] }, "vague": { "episodes": 1200, "successes": 185, "sr": 15.4, "score": null, "score_ci_bootstrap": null, "ee_speed_cm_s": 6.35, "ee_speed_sd_cm_s": 1.74, "ee_sparc": -5.82, "by_difficulty": { "simple": { "episodes": 640, "successes": 112, "sr": 17.5, "score": null, "score_ci_bootstrap": null, "sr_ci_cp": [ 14.633569253956317, 20.669885557199237 ] }, "moderate": { "episodes": 390, "successes": 62, "sr": 15.9, "score": null, "score_ci_bootstrap": null, "sr_ci_cp": [ 12.410935937399675, 19.911350571272386 ] }, "complex": { "episodes": 170, "successes": 11, "sr": 6.5, "score": null, "score_ci_bootstrap": null, "sr_ci_cp": [ 3.274070258867132, 11.282581642664729 ] } }, "by_axis": { "procedural": { "episodes": 220, "successes": 26, "sr": 11.8, "score": null, "score_ci_bootstrap": null, "sr_ci_cp": [ 7.867108772663284, 16.835298357182577 ] }, "relational": { "episodes": 420, "successes": 98, "sr": 23.3, "score": null, "score_ci_bootstrap": null, "sr_ci_cp": [ 19.36945924388729, 27.67752468108109 ] }, "visual": { "episodes": 840, "successes": 97, "sr": 11.5, "score": null, "score_ci_bootstrap": null, "sr_ci_cp": [ 9.464748342203, 13.904524539896546 ] } }, "episodes_per_task": "10", "sr_ci_cp": [ 13.418562910201498, 17.585850392735118 ] }, "specific": { "episodes": 1200, "successes": 346, "sr": 28.8, "score": null, "score_ci_bootstrap": null, "ee_speed_cm_s": 7.14, "ee_speed_sd_cm_s": 1.89, "ee_sparc": -5.43, "by_difficulty": { "simple": { "episodes": 640, "successes": 204, "sr": 31.9, "score": null, "score_ci_bootstrap": null, "sr_ci_cp": [ 28.27735390927054, 35.640850122061735 ] }, "moderate": { "episodes": 390, "successes": 117, "sr": 30.0, "score": null, "score_ci_bootstrap": null, "sr_ci_cp": [ 25.49065937332436, 34.81597009792536 ] }, "complex": { "episodes": 170, "successes": 25, "sr": 14.7, "score": null, "score_ci_bootstrap": null, "sr_ci_cp": [ 9.74852841315704, 20.9388600948618 ] } }, "by_axis": { "procedural": { "episodes": 220, "successes": 38, "sr": 17.3, "score": null, "score_ci_bootstrap": null, "sr_ci_cp": [ 12.52261312732176, 22.928411609027197 ] }, "relational": { "episodes": 420, "successes": 144, "sr": 34.3, "score": null, "score_ci_bootstrap": null, "sr_ci_cp": [ 29.751872921101942, 39.042786717835256 ] }, "visual": { "episodes": 840, "successes": 195, "sr": 23.2, "score": null, "score_ci_bootstrap": null, "sr_ci_cp": [ 20.397711411256736, 26.220165302743947 ] } }, "episodes_per_task": "10", "sr_ci_cp": [ 26.283161344643663, 31.487667403692065 ] } }, "badges": [ "Organizer cross-check" ], "links": [ { "label": "Project page", "url": "https://huggingface.co/nvidia/Cosmos3-Edge-Policy-DROID", "kind": "project" } ], "table": { "observations": [ "RGB", "P" ], "vram": "9.20 GB" } }, { "id": "pi0-fast", "display_name": "π0-FAST", "organization": "Physical Intelligence", "organization_links": [ { "label": "Physical Intelligence", "url": "https://www.pi.website/" } ], "source": "internal", "policy": { "type": "VLA", "type_other": null, "description": "An autoregressive vision-language-action policy that uses compressed action tokens to efficiently produce dexterous, high-frequency robot behavior.", "robot": "DROID (Franka)", "observation_space": [ "RGB", "Proprioception" ], "action_space": [ "Joint position" ], "action_horizon_steps": "10 steps", "size": "3B", "gpu": "NVIDIA L40S", "vram": ">8 GB (OpenPI inference requirement)", "cameras": [ "EGO-L", "Wrist" ], "observation_history_label": "Current" }, "open_sourcing": null, "training": { "sim_image_data": { "robolab_in_finetuning": false, "robolab_in_training": false, "other_sim_in_training": false } }, "provenance": { "display_date": "Evaluated May 9, 2026", "robolab_build": "09545754 (v0.1.0)", "isaac_versions": "2.2.0 / 5.0.0" }, "results": { "default": { "episodes": 1200, "successes": 186, "sr": 15.5, "score": 26.9, "score_ci_bootstrap": [ 21.0, 32.7 ], "ee_speed_cm_s": 4.6, "ee_speed_sd_cm_s": 1.77, "ee_sparc": -9.63, "by_difficulty": { "simple": { "episodes": 640, "successes": 129, "sr": 20.2, "score": 26.7, "score_ci_bootstrap": [ 18.4, 35.6 ], "sr_ci_cp": [ 17.113678272922836, 23.476990544427633 ] }, "moderate": { "episodes": 390, "successes": 52, "sr": 13.3, "score": 29.2, "score_ci_bootstrap": [ 18.9, 40.5 ], "sr_ci_cp": [ 10.120753537351417, 17.114808557392237 ] }, "complex": { "episodes": 170, "successes": 5, "sr": 2.9, "score": 21.9, "score_ci_bootstrap": [ 12.5, 32.4 ], "sr_ci_cp": [ 0.9617497723282539, 6.729969179077931 ] } }, "by_axis": { "procedural": { "episodes": 220, "successes": 14, "sr": 6.4, "score": 16.4, "score_ci_bootstrap": [ 8.3, 25.6 ], "sr_ci_cp": [ 3.52241133156329, 10.446677161316776 ] }, "relational": { "episodes": 420, "successes": 98, "sr": 23.3, "score": 36.7, "score_ci_bootstrap": [ 25.3, 48.6 ], "sr_ci_cp": [ 19.36945924388729, 27.67752468108109 ] }, "visual": { "episodes": 840, "successes": 74, "sr": 8.8, "score": 19.5, "score_ci_bootstrap": [ 13.6, 25.9 ], "sr_ci_cp": [ 6.980535116615738, 10.933330929522853 ] } }, "episodes_per_task": "10", "sr_ci_cp": [ 13.497191830061583, 17.673466253407284 ] }, "vague": { "episodes": 1200, "successes": 111, "sr": 9.2, "score": 16.8, "score_ci_bootstrap": [ 11.6, 22.2 ], "ee_speed_cm_s": 4.85, "ee_speed_sd_cm_s": 1.77, "ee_sparc": -9.9, "by_difficulty": { "simple": { "episodes": 640, "successes": 61, "sr": 9.5, "score": 13.0, "score_ci_bootstrap": [ 6.5, 20.8 ], "sr_ci_cp": [ 7.369154494066124, 12.07483270531895 ] }, "moderate": { "episodes": 390, "successes": 50, "sr": 12.8, "score": 25.6, "score_ci_bootstrap": [ 15.6, 36.3 ], "sr_ci_cp": [ 9.667407593004693, 16.55091093466967 ] }, "complex": { "episodes": 170, "successes": 0, "sr": 0.0, "score": 10.4, "score_ci_bootstrap": [ 5.0, 16.4 ], "sr_ci_cp": [ 0.0, 2.1465554980944557 ] } }, "by_axis": { "procedural": { "episodes": 220, "successes": 7, "sr": 3.2, "score": 14.2, "score_ci_bootstrap": [ 7.7, 21.5 ], "sr_ci_cp": [ 1.2886409653742155, 6.445604600496031 ] }, "relational": { "episodes": 420, "successes": 79, "sr": 18.8, "score": 29.4, "score_ci_bootstrap": [ 18.3, 40.6 ], "sr_ci_cp": [ 15.184106961059216, 22.88109643172638 ] }, "visual": { "episodes": 840, "successes": 44, "sr": 5.2, "score": 11.6, "score_ci_bootstrap": [ 6.8, 17.1 ], "sr_ci_cp": [ 3.831456185043743, 6.968403312306036 ] } }, "episodes_per_task": "10", "sr_ci_cp": [ 7.670491200138851, 11.032365580535682 ] }, "specific": { "episodes": 1200, "successes": 179, "sr": 14.9, "score": 26.9, "score_ci_bootstrap": [ 20.9, 33.0 ], "ee_speed_cm_s": 4.66, "ee_speed_sd_cm_s": 1.77, "ee_sparc": -9.48, "by_difficulty": { "simple": { "episodes": 640, "successes": 124, "sr": 19.4, "score": 24.8, "score_ci_bootstrap": [ 16.6, 33.9 ], "sr_ci_cp": [ 16.381886272154638, 22.65369826676408 ] }, "moderate": { "episodes": 390, "successes": 49, "sr": 12.6, "score": 31.3, "score_ci_bootstrap": [ 20.8, 42.4 ], "sr_ci_cp": [ 9.441383699098473, 16.268328398841174 ] }, "complex": { "episodes": 170, "successes": 6, "sr": 3.5, "score": 25.1, "score_ci_bootstrap": [ 15.1, 36.5 ], "sr_ci_cp": [ 1.3060497433482048, 7.523461385455331 ] } }, "by_axis": { "procedural": { "episodes": 220, "successes": 9, "sr": 4.1, "score": 16.0, "score_ci_bootstrap": [ 6.8, 26.6 ], "sr_ci_cp": [ 1.8874184171802662, 7.6235228673933255 ] }, "relational": { "episodes": 420, "successes": 87, "sr": 20.7, "score": 34.9, "score_ci_bootstrap": [ 23.4, 46.4 ], "sr_ci_cp": [ 16.93750893577969, 24.909390591864117 ] }, "visual": { "episodes": 840, "successes": 84, "sr": 10.0, "score": 21.2, "score_ci_bootstrap": [ 14.9, 28.0 ], "sr_ci_cp": [ 8.05491849532462, 12.230777967873637 ] } }, "episodes_per_task": "10", "sr_ci_cp": [ 12.947232251286847, 17.059715246437133 ] } }, "badges": [ "Organizer cross-check" ], "links": [ { "label": "Project page", "url": "https://www.physicalintelligence.company/research/fast", "kind": "project" } ], "table": { "observations": [ "RGB", "P" ], "vram": ">8 GB" } }, { "id": "gr00t", "display_name": "GR00T N1.6", "organization": "NVIDIA GEAR", "organization_links": [ { "label": "NVIDIA GEAR", "url": "https://research.nvidia.com/labs/gear/" } ], "source": "internal", "policy": { "type": "VLA", "type_other": null, "description": "A multimodal VLA foundation model combining camera observations, robot state, and language instructions for cross-embodiment robot control.", "robot": "DROID (Franka)", "observation_space": [ "RGB", "Proprioception" ], "action_space": [ "Joint position" ], "action_horizon_steps": "10 steps", "size": "3B", "gpu": "NVIDIA L40S", "vram": "24 GB (RTX 4090 public reference setup)", "cameras": [ "EGO-L", "Wrist" ], "observation_history_label": "Current" }, "open_sourcing": null, "training": { "sim_image_data": { "robolab_in_finetuning": false, "robolab_in_training": false, "other_sim_in_training": false } }, "provenance": { "display_date": "Evaluated May 9, 2026", "robolab_build": "09545754 (v0.1.0)", "isaac_versions": "2.2.0 / 5.0.0" }, "results": { "default": { "episodes": 1200, "successes": 87, "sr": 7.2, "score": 17.1, "score_ci_bootstrap": [ 12.4, 22.1 ], "ee_speed_cm_s": 4.3, "ee_speed_sd_cm_s": 1.41, "ee_sparc": -6.87, "by_difficulty": { "simple": { "episodes": 640, "successes": 56, "sr": 8.8, "score": 13.6, "score_ci_bootstrap": [ 8.0, 19.9 ], "sr_ci_cp": [ 6.6772341915106175, 11.212251092837215 ] }, "moderate": { "episodes": 390, "successes": 31, "sr": 7.9, "score": 25.1, "score_ci_bootstrap": [ 14.9, 36.2 ], "sr_ci_cp": [ 5.464475891830321, 11.092956816934498 ] }, "complex": { "episodes": 170, "successes": 0, "sr": 0.0, "score": 11.9, "score_ci_bootstrap": [ 6.6, 17.8 ], "sr_ci_cp": [ 0.0, 2.1465554980944557 ] } }, "by_axis": { "procedural": { "episodes": 220, "successes": 6, "sr": 2.7, "score": 11.1, "score_ci_bootstrap": [ 3.8, 20.5 ], "sr_ci_cp": [ 1.007298622288058, 5.841310210583319 ] }, "relational": { "episodes": 420, "successes": 42, "sr": 10.0, "score": 23.2, "score_ci_bootstrap": [ 13.7, 33.6 ], "sr_ci_cp": [ 7.302528234982003, 13.276312315959668 ] }, "visual": { "episodes": 840, "successes": 42, "sr": 5.0, "score": 15.1, "score_ci_bootstrap": [ 9.9, 20.9 ], "sr_ci_cp": [ 3.6269828384132086, 6.698748874224273 ] } }, "episodes_per_task": "10", "sr_ci_cp": [ 5.847382337225042, 8.866217103560253 ] }, "vague": { "episodes": 1200, "successes": 65, "sr": 5.4, "score": 14.0, "score_ci_bootstrap": [ 9.7, 18.7 ], "ee_speed_cm_s": 4.19, "ee_speed_sd_cm_s": 1.24, "ee_sparc": -6.61, "by_difficulty": { "simple": { "episodes": 640, "successes": 46, "sr": 7.2, "score": 9.9, "score_ci_bootstrap": [ 5.2, 15.5 ], "sr_ci_cp": [ 5.309884653081947, 9.470985729554922 ] }, "moderate": { "episodes": 390, "successes": 19, "sr": 4.9, "score": 22.8, "score_ci_bootstrap": [ 13.5, 33.1 ], "sr_ci_cp": [ 2.9582738579716517, 7.503576393721634 ] }, "complex": { "episodes": 170, "successes": 0, "sr": 0.0, "score": 9.0, "score_ci_bootstrap": [ 3.6, 15.4 ], "sr_ci_cp": [ 0.0, 2.1465554980944557 ] } }, "by_axis": { "procedural": { "episodes": 220, "successes": 9, "sr": 4.1, "score": 15.2, "score_ci_bootstrap": [ 6.8, 25.2 ], "sr_ci_cp": [ 1.8874184171802662, 7.6235228673933255 ] }, "relational": { "episodes": 420, "successes": 20, "sr": 4.8, "score": 19.6, "score_ci_bootstrap": [ 11.3, 29.0 ], "sr_ci_cp": [ 2.9326134463751026, 7.258790787247772 ] }, "visual": { "episodes": 840, "successes": 36, "sr": 4.3, "score": 12.3, "score_ci_bootstrap": [ 7.7, 17.4 ], "sr_ci_cp": [ 3.0194147616340428, 5.8841227587662965 ] } }, "episodes_per_task": "10", "sr_ci_cp": [ 4.205008247519013, 6.852216385165553 ] }, "specific": { "episodes": 1200, "successes": 64, "sr": 5.3, "score": 17.1, "score_ci_bootstrap": [ 12.8, 21.9 ], "ee_speed_cm_s": 4.45, "ee_speed_sd_cm_s": 1.41, "ee_sparc": -6.84, "by_difficulty": { "simple": { "episodes": 640, "successes": 48, "sr": 7.5, "score": 14.1, "score_ci_bootstrap": [ 8.5, 20.9 ], "sr_ci_cp": [ 5.581388346098379, 9.82115458544903 ] }, "moderate": { "episodes": 390, "successes": 16, "sr": 4.1, "score": 23.3, "score_ci_bootstrap": [ 14.4, 32.9 ], "sr_ci_cp": [ 2.362853697947943, 6.5769413442232 ] }, "complex": { "episodes": 170, "successes": 0, "sr": 0.0, "score": 13.9, "score_ci_bootstrap": [ 9.1, 18.8 ], "sr_ci_cp": [ 0.0, 2.1465554980944557 ] } }, "by_axis": { "procedural": { "episodes": 220, "successes": 8, "sr": 3.6, "score": 12.3, "score_ci_bootstrap": [ 4.5, 22.9 ], "sr_ci_cp": [ 1.5827676670189157, 7.039095643278634 ] }, "relational": { "episodes": 420, "successes": 26, "sr": 6.2, "score": 22.8, "score_ci_bootstrap": [ 14.6, 31.8 ], "sr_ci_cp": [ 4.0833104368629565, 8.939176752544418 ] }, "visual": { "episodes": 840, "successes": 23, "sr": 2.7, "score": 14.7, "score_ci_bootstrap": [ 10.3, 19.6 ], "sr_ci_cp": [ 1.7434619592852512, 4.080281770015104 ] } }, "episodes_per_task": "10", "sr_ci_cp": [ 4.1312073587989735, 6.759834309392418 ] } }, "badges": [ "Organizer cross-check" ], "links": [ { "label": "Project page", "url": "https://developer.nvidia.com/isaac/gr00t", "kind": "project" } ], "table": { "observations": [ "RGB", "P" ], "vram": "24 GB" } }, { "id": "pi0", "display_name": "π0", "organization": "Physical Intelligence", "organization_links": [ { "label": "Physical Intelligence", "url": "https://www.pi.website/" } ], "source": "internal", "policy": { "type": "VLA", "type_other": null, "description": "A generalist VLA model that turns images, language, and robot state into continuous low-level actions across multiple robot embodiments.", "robot": "DROID (Franka)", "observation_space": [ "RGB", "Proprioception" ], "action_space": [ "Joint position" ], "action_horizon_steps": "10 steps", "size": "3.3B", "gpu": "NVIDIA L40S", "vram": ">8 GB (OpenPI inference requirement)", "cameras": [ "EGO-L", "Wrist" ], "observation_history_label": "Current" }, "open_sourcing": null, "training": { "sim_image_data": { "robolab_in_finetuning": false, "robolab_in_training": false, "other_sim_in_training": false } }, "provenance": { "display_date": "Evaluated May 9, 2026", "robolab_build": "09545754 (v0.1.0)", "isaac_versions": "2.2.0 / 5.0.0" }, "results": { "default": { "episodes": 1200, "successes": 60, "sr": 5.0, "score": 12.2, "score_ci_bootstrap": [ 8.5, 16.4 ], "ee_speed_cm_s": 4.18, "ee_speed_sd_cm_s": 1.24, "ee_sparc": -9.49, "by_difficulty": { "simple": { "episodes": 640, "successes": 46, "sr": 7.2, "score": 9.6, "score_ci_bootstrap": [ 4.9, 15.2 ], "sr_ci_cp": [ 5.309884653081947, 9.470985729554922 ] }, "moderate": { "episodes": 390, "successes": 14, "sr": 3.6, "score": 18.1, "score_ci_bootstrap": [ 10.3, 27.0 ], "sr_ci_cp": [ 1.9761885311082767, 5.949680879379189 ] }, "complex": { "episodes": 170, "successes": 0, "sr": 0.0, "score": 8.7, "score_ci_bootstrap": [ 3.8, 14.1 ], "sr_ci_cp": [ 0.0, 2.1465554980944557 ] } }, "by_axis": { "procedural": { "episodes": 220, "successes": 1, "sr": 0.5, "score": 6.6, "score_ci_bootstrap": [ 2.2, 12.4 ], "sr_ci_cp": [ 0.011507432382514825, 2.506407622796453 ] }, "relational": { "episodes": 420, "successes": 27, "sr": 6.4, "score": 18.8, "score_ci_bootstrap": [ 11.2, 27.3 ], "sr_ci_cp": [ 4.278855182409012, 9.215674414174538 ] }, "visual": { "episodes": 840, "successes": 16, "sr": 1.9, "score": 8.2, "score_ci_bootstrap": [ 4.7, 12.3 ], "sr_ci_cp": [ 1.0925580939215302, 3.074824075134678 ] } }, "episodes_per_task": "10", "sr_ci_cp": [ 3.836891443711108, 6.389437121473936 ] }, "vague": { "episodes": 1200, "successes": 33, "sr": 2.8, "score": 8.5, "score_ci_bootstrap": [ 5.1, 12.2 ], "ee_speed_cm_s": 4.24, "ee_speed_sd_cm_s": 1.24, "ee_sparc": -10.36, "by_difficulty": { "simple": { "episodes": 640, "successes": 18, "sr": 2.8, "score": 3.9, "score_ci_bootstrap": [ 0.9, 8.0 ], "sr_ci_cp": [ 1.6752006298100315, 4.408638286600256 ] }, "moderate": { "episodes": 390, "successes": 15, "sr": 3.8, "score": 17.0, "score_ci_bootstrap": [ 8.9, 25.9 ], "sr_ci_cp": [ 2.16836501479302, 6.264373221964159 ] }, "complex": { "episodes": 170, "successes": 0, "sr": 0.0, "score": 6.0, "score_ci_bootstrap": [ 1.6, 11.3 ], "sr_ci_cp": [ 0.0, 2.1465554980944557 ] } }, "by_axis": { "procedural": { "episodes": 220, "successes": 3, "sr": 1.4, "score": 6.8, "score_ci_bootstrap": [ 1.6, 13.9 ], "sr_ci_cp": [ 0.2821022949031471, 3.933186817676205 ] }, "relational": { "episodes": 420, "successes": 26, "sr": 6.2, "score": 18.5, "score_ci_bootstrap": [ 10.4, 27.7 ], "sr_ci_cp": [ 4.0833104368629565, 8.939176752544418 ] }, "visual": { "episodes": 840, "successes": 7, "sr": 0.8, "score": 5.2, "score_ci_bootstrap": [ 2.3, 8.9 ], "sr_ci_cp": [ 0.3356807180589102, 1.7094070078856456 ] } }, "episodes_per_task": "10", "sr_ci_cp": [ 1.9003686091546828, 3.8404873460767286 ] }, "specific": { "episodes": 1200, "successes": 42, "sr": 3.5, "score": 10.3, "score_ci_bootstrap": [ 7.0, 14.2 ], "ee_speed_cm_s": 4.12, "ee_speed_sd_cm_s": 1.24, "ee_sparc": -9.21, "by_difficulty": { "simple": { "episodes": 640, "successes": 34, "sr": 5.3, "score": 7.8, "score_ci_bootstrap": [ 3.5, 13.1 ], "sr_ci_cp": [ 3.706757269036778, 7.344868221040706 ] }, "moderate": { "episodes": 390, "successes": 8, "sr": 2.1, "score": 15.3, "score_ci_bootstrap": [ 8.7, 22.9 ], "sr_ci_cp": [ 0.8896582807653726, 4.0016931583331505 ] }, "complex": { "episodes": 170, "successes": 0, "sr": 0.0, "score": 8.5, "score_ci_bootstrap": [ 4.1, 13.5 ], "sr_ci_cp": [ 0.0, 2.1465554980944557 ] } }, "by_axis": { "procedural": { "episodes": 220, "successes": 2, "sr": 0.9, "score": 6.7, "score_ci_bootstrap": [ 1.8, 13.2 ], "sr_ci_cp": [ 0.11028536311874819, 3.2451847043022966 ] }, "relational": { "episodes": 420, "successes": 17, "sr": 4.0, "score": 15.8, "score_ci_bootstrap": [ 9.0, 23.4 ], "sr_ci_cp": [ 2.3752705787109645, 6.401679137367671 ] }, "visual": { "episodes": 840, "successes": 13, "sr": 1.5, "score": 7.6, "score_ci_bootstrap": [ 4.3, 11.4 ], "sr_ci_cp": [ 0.8265479577196071, 2.6319360980949247 ] } }, "episodes_per_task": "10", "sr_ci_cp": [ 2.5339083209800486, 4.701737881740953 ] } }, "badges": [ "Organizer cross-check" ], "links": [ { "label": "Project page", "url": "https://www.physicalintelligence.company/blog/pi0", "kind": "project" } ], "table": { "observations": [ "RGB", "P" ], "vram": ">8 GB" } }, { "id": "paligemma", "display_name": "paligemma-binning", "organization": "Physical Intelligence", "organization_links": [ { "label": "Physical Intelligence", "url": "https://www.pi.website/" } ], "source": "internal", "policy": { "type": "VLA", "type_other": null, "description": "A PaliGemma-based vision-language-action baseline that predicts discretized joint-action bins from language-conditioned visual and proprioceptive observations.", "robot": "DROID (Franka)", "observation_space": [ "RGB", "Proprioception" ], "action_space": [ "Joint position" ], "action_horizon_steps": "10 steps", "size": "3B", "gpu": "NVIDIA L40S", "vram": ">8 GB (OpenPI inference requirement)", "cameras": [ "EGO-L", "Wrist" ], "observation_history_label": "Current" }, "open_sourcing": null, "training": { "sim_image_data": { "robolab_in_finetuning": false, "robolab_in_training": false, "other_sim_in_training": false } }, "provenance": { "display_date": "Evaluated May 9, 2026", "robolab_build": "09545754 (v0.1.0)", "isaac_versions": "2.2.0 / 5.0.0" }, "results": { "default": { "episodes": 1200, "successes": 41, "sr": 3.4, "score": 9.9, "score_ci_bootstrap": [ 6.5, 13.7 ], "ee_speed_cm_s": 1.93, "ee_speed_sd_cm_s": 1.59, "ee_sparc": -16.52, "by_difficulty": { "simple": { "episodes": 640, "successes": 22, "sr": 3.4, "score": 6.2, "score_ci_bootstrap": [ 2.5, 10.6 ], "sr_ci_cp": [ 2.1665592913922995, 5.158333019146778 ] }, "moderate": { "episodes": 390, "successes": 19, "sr": 4.9, "score": 16.2, "score_ci_bootstrap": [ 8.9, 24.4 ], "sr_ci_cp": [ 2.9582738579716517, 7.503576393721634 ] }, "complex": { "episodes": 170, "successes": 0, "sr": 0.0, "score": 9.0, "score_ci_bootstrap": [ 3.4, 15.7 ], "sr_ci_cp": [ 0.0, 2.1465554980944557 ] } }, "by_axis": { "procedural": { "episodes": 220, "successes": 1, "sr": 0.5, "score": 4.0, "score_ci_bootstrap": [ 0.9, 8.2 ], "sr_ci_cp": [ 0.011507432382514825, 2.506407622796453 ] }, "relational": { "episodes": 420, "successes": 24, "sr": 5.7, "score": 15.6, "score_ci_bootstrap": [ 8.4, 23.7 ], "sr_ci_cp": [ 3.6951925187137893, 8.38335491599133 ] }, "visual": { "episodes": 840, "successes": 23, "sr": 2.7, "score": 9.3, "score_ci_bootstrap": [ 5.3, 14.0 ], "sr_ci_cp": [ 1.7434619592852512, 4.080281770015104 ] } }, "episodes_per_task": "10", "sr_ci_cp": [ 2.4627950157516167, 4.606738475713753 ] }, "vague": { "episodes": 1200, "successes": 37, "sr": 3.1, "score": 8.6, "score_ci_bootstrap": [ 5.1, 12.6 ], "ee_speed_cm_s": 1.98, "ee_speed_sd_cm_s": 1.77, "ee_sparc": -17.89, "by_difficulty": { "simple": { "episodes": 640, "successes": 14, "sr": 2.2, "score": 4.2, "score_ci_bootstrap": [ 1.1, 8.1 ], "sr_ci_cp": [ 1.2009644619712003, 3.6430383294670263 ] }, "moderate": { "episodes": 390, "successes": 23, "sr": 5.9, "score": 16.6, "score_ci_bootstrap": [ 8.6, 25.9 ], "sr_ci_cp": [ 3.774828463978832, 8.717896087148349 ] }, "complex": { "episodes": 170, "successes": 0, "sr": 0.0, "score": 6.5, "score_ci_bootstrap": [ 1.5, 12.5 ], "sr_ci_cp": [ 0.0, 2.1465554980944557 ] } }, "by_axis": { "procedural": { "episodes": 220, "successes": 2, "sr": 0.9, "score": 4.2, "score_ci_bootstrap": [ 0.9, 8.7 ], "sr_ci_cp": [ 0.11028536311874819, 3.2451847043022966 ] }, "relational": { "episodes": 420, "successes": 24, "sr": 5.7, "score": 15.4, "score_ci_bootstrap": [ 7.5, 24.6 ], "sr_ci_cp": [ 3.6951925187137893, 8.38335491599133 ] }, "visual": { "episodes": 840, "successes": 30, "sr": 3.6, "score": 9.2, "score_ci_bootstrap": [ 4.7, 14.4 ], "sr_ci_cp": [ 2.422354483097392, 5.059374782719494 ] } }, "episodes_per_task": "10", "sr_ci_cp": [ 2.180048794102941, 4.2250906917481625 ] }, "specific": { "episodes": 1200, "successes": 66, "sr": 5.5, "score": 11.8, "score_ci_bootstrap": [ 8.0, 15.9 ], "ee_speed_cm_s": 2.01, "ee_speed_sd_cm_s": 1.77, "ee_sparc": -16.3, "by_difficulty": { "simple": { "episodes": 640, "successes": 26, "sr": 4.1, "score": 6.6, "score_ci_bootstrap": [ 3.0, 10.9 ], "sr_ci_cp": [ 2.670629403757821, 5.89606086490904 ] }, "moderate": { "episodes": 390, "successes": 40, "sr": 10.3, "score": 20.6, "score_ci_bootstrap": [ 11.9, 30.3 ], "sr_ci_cp": [ 7.428920882362846, 13.7039020169231 ] }, "complex": { "episodes": 170, "successes": 0, "sr": 0.0, "score": 10.7, "score_ci_bootstrap": [ 5.3, 16.8 ], "sr_ci_cp": [ 0.0, 2.1465554980944557 ] } }, "by_axis": { "procedural": { "episodes": 220, "successes": 9, "sr": 4.1, "score": 8.4, "score_ci_bootstrap": [ 2.4, 16.4 ], "sr_ci_cp": [ 1.8874184171802662, 7.6235228673933255 ] }, "relational": { "episodes": 420, "successes": 40, "sr": 9.5, "score": 19.2, "score_ci_bootstrap": [ 10.9, 28.6 ], "sr_ci_cp": [ 6.891307301757571, 12.74267652586381 ] }, "visual": { "episodes": 840, "successes": 42, "sr": 5.0, "score": 10.6, "score_ci_bootstrap": [ 6.3, 15.4 ], "sr_ci_cp": [ 3.6269828384132086, 6.698748874224273 ] } }, "episodes_per_task": "10", "sr_ci_cp": [ 4.278894011958297, 6.944515321103239 ] } }, "badges": [ "Organizer cross-check" ], "links": [ { "label": "Project page", "url": "https://arxiv.org/abs/2407.07726", "kind": "project" } ], "table": { "observations": [ "RGB", "P" ], "vram": ">8 GB" } } ] }