diff --git a/masters/embodiedswe/assembly_allen_bolt_franka_osc/meta.json b/masters/embodiedswe/assembly_allen_bolt_franka_osc/meta.json new file mode 100644 index 0000000000000000000000000000000000000000..1a8b7c940e90741a49d2832519a459844a6c3dec --- /dev/null +++ b/masters/embodiedswe/assembly_allen_bolt_franka_osc/meta.json @@ -0,0 +1,52 @@ +{ + "suite": "embodiedswe", + "task": "assembly_allen_bolt_franka_osc", + "task_dir": "embodiedswe-assembly-allen-bolt-franka-osc-i00-standard", + "source": "agent-success:codex-gpt6_luna-xhigh-embodiedswe-assembly-allen-bolt-franka-osc-i00-standard/embodiedswe-assembly-allen-bolt__AaZsDRG", + "agent": "Codex + GPT-6 Luna (xhigh)", + "run": "codex-gpt6_luna-xhigh", + "mode": "standard", + "trial": "AaZsDRG", + "trial_choice": "the only run that solved this task (GPT-6 Luna, counted standard success); the other runs did not attempt it (screening: they run only tasks Luna fails in every mode)", + "renderer": "isaac-rtx PathTracing spp=32 totalSpp=32 maxBounces=6 optix-denoiser firefly-filter (Isaac Sim 5.1 / Isaac Lab 2.3.2; dedicated render camera at the scene's own viewpoint, rendered natively at 2560x1440, focal 18.0 mm; recorded states posed, render only)", + "rendering": "path-tracing", + "width": 2560, + "height": 1440, + "fps": 30, + "frames": 786, + "seconds": 26.2, + "speed": 32.0, + "capture": "one recorded state every 16 control steps (1.067 s of simulated time) per video frame = 32x real time, 726 rendered frames; 0.5 s hold of the first frame, 1.5 s hold of the final frame", + "camera": { + "eye": [ + 1.1, + -0.72, + 0.85 + ], + "target": [ + 0.36, + 0.05, + 0.2 + ], + "focal_mm": 18.0, + "room_camera": [ + [ + 1.1, + -0.72, + 0.85 + ], + [ + 0.36, + 0.05, + 0.2 + ] + ] + }, + "success_check": true, + "success_check_how": "the video poses the graded episode's own recorded scene states (trajectory.npz frame_state: every asset's root pose and velocity and joint state, recorded during the graded live standard episode by the simulator service) on the task's own scene build (InteractiveScene.reset_to, no physics step), so it shows the graded run itself. Start state = the frozen instance (instance_ok 1, NVIDIA L40S). The task's success predicate on the posed final state: 1. Verdict (reward.json): success 1, progress 1.0.", + "image": "efe-private:embodiedswe-0.3.0 (sim service: efe:embodiedswe-standard-files-0.2.0)", + "core": false, + "encode": "libx264 preset slow CRF 16, yuv420p, bt709, +faststart", + "poster": "the final frame", + "created": "2026-10-30" +} \ No newline at end of file diff --git a/masters/embodiedswe/assembly_allen_bolt_franka_osc/poster.png b/masters/embodiedswe/assembly_allen_bolt_franka_osc/poster.png new file mode 100644 index 0000000000000000000000000000000000000000..117c302d6bd9e66b1bf23a9f756492d622c4fca0 --- /dev/null +++ b/masters/embodiedswe/assembly_allen_bolt_franka_osc/poster.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0e797b76516a071b7f5369e00a8bc6df23eca2a8d8e094918881da0fb30dbcfb +size 2894567 diff --git a/masters/embodiedswe/assembly_allen_bolt_franka_osc/video.mp4 b/masters/embodiedswe/assembly_allen_bolt_franka_osc/video.mp4 new file mode 100644 index 0000000000000000000000000000000000000000..87215b3a37ed7d4aa4b613128806c5ed74720e18 --- /dev/null +++ b/masters/embodiedswe/assembly_allen_bolt_franka_osc/video.mp4 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1a88e216eef9232b15c4c6a9c35d6129ef0c6068a12fd2e59f8a86da02da3399 +size 16549848 diff --git a/masters/embodiedswe/assembly_bulb_franka_osc/meta.json b/masters/embodiedswe/assembly_bulb_franka_osc/meta.json new file mode 100644 index 0000000000000000000000000000000000000000..7b0fe7f3c650869ffb33019c5f688afee69ff959 --- /dev/null +++ b/masters/embodiedswe/assembly_bulb_franka_osc/meta.json @@ -0,0 +1,52 @@ +{ + "suite": "embodiedswe", + "task": "assembly_bulb_franka_osc", + "task_dir": "embodiedswe-assembly-bulb-franka-osc-i00-privileged", + "source": "agent-success:codex-gpt6_astra-high-embodiedswe-assembly-bulb-franka-osc-i00-privileged/embodiedswe-assembly-bulb-franka__KWguxBe", + "agent": "Codex + GPT-6 Astra (high)", + "run": "codex-gpt6_astra-high", + "mode": "privileged", + "trial": "KWguxBe", + "trial_choice": "counted privileged success; order: GPT-6 Astra > Claude Opus 5.5 > Gemini 4 Argon > Gemini 3.8 Flash, then the task's default mode (standard) before privileged", + "renderer": "isaac-rtx PathTracing spp=32 totalSpp=32 maxBounces=6 optix-denoiser firefly-filter (Isaac Sim 5.1 / Isaac Lab 2.3.2; dedicated render camera at the scene's own viewpoint, rendered natively at 2560x1440, focal 18.0 mm; recorded states posed, render only)", + "rendering": "path-tracing", + "width": 2560, + "height": 1440, + "fps": 30, + "frames": 623, + "seconds": 20.77, + "speed": 8.0, + "capture": "one recorded state every 4 control steps (0.267 s of simulated time) per video frame = 8x real time, 563 rendered frames; 0.5 s hold of the first frame, 1.5 s hold of the final frame", + "camera": { + "eye": [ + 1.1, + -0.72, + 0.85 + ], + "target": [ + 0.36, + 0.05, + 0.2 + ], + "focal_mm": 18.0, + "room_camera": [ + [ + 1.1, + -0.72, + 0.85 + ], + [ + 0.36, + 0.05, + 0.2 + ] + ] + }, + "success_check": true, + "success_check_how": "the video poses the graded episode's own recorded scene states (trajectory.npz frame_state: every asset's root pose and velocity and joint state, recorded during the graded verifier rollout) on the task's own scene build (InteractiveScene.reset_to, no physics step), so it shows the graded run itself. Start state = the frozen instance (instance_ok 1, NVIDIA L40S). The task's success predicate on the posed final state: 1. Verdict (reward.json): success 1, progress 1.0. The verifier's own camera replay of the actions ended on the graded final state: camera_replay_match 1.", + "image": "efe-private:embodiedswe-0.3.0 (verifier tooling: efe:embodiedswe-files-0.3.1)", + "core": true, + "encode": "libx264 preset slow CRF 16, yuv420p, bt709, +faststart", + "poster": "the final frame", + "created": "2026-10-30" +} \ No newline at end of file diff --git a/masters/embodiedswe/assembly_bulb_franka_osc/poster.png b/masters/embodiedswe/assembly_bulb_franka_osc/poster.png new file mode 100644 index 0000000000000000000000000000000000000000..ca3d815957251269a041af33dedb4967127c6d5e --- /dev/null +++ b/masters/embodiedswe/assembly_bulb_franka_osc/poster.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:302270a50d573f30ace6e893569e468dce250bb2cc25f1ff768c596ced1fdd03 +size 2860900 diff --git a/masters/embodiedswe/assembly_bulb_franka_osc/video.mp4 b/masters/embodiedswe/assembly_bulb_franka_osc/video.mp4 new file mode 100644 index 0000000000000000000000000000000000000000..dc4ede42abd0a070e83996c98df02101fc6de518 --- /dev/null +++ b/masters/embodiedswe/assembly_bulb_franka_osc/video.mp4 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7d7bc37b521100971112c4fdf8676cfcaf730aea90cd9ed80ad8445e908bc118 +size 8349546 diff --git a/masters/embodiedswe/assembly_ikea_table_bimanual_franka_osc/meta.json b/masters/embodiedswe/assembly_ikea_table_bimanual_franka_osc/meta.json new file mode 100644 index 0000000000000000000000000000000000000000..e3deda5624eb84fb0da1eca9473740533c38cbff --- /dev/null +++ b/masters/embodiedswe/assembly_ikea_table_bimanual_franka_osc/meta.json @@ -0,0 +1,54 @@ +{ + "suite": "embodiedswe", + "task": "assembly_ikea_table_bimanual_franka_osc", + "task_dir": "embodiedswe-assembly-ikea-table-bimanual-franka-osc-i00-privileged", + "source": "scene-orbit", + "agent": "Codex + GPT-6 Astra (high)", + "run": "codex-gpt6_astra-high", + "mode": "privileged", + "trial": "464xuPn", + "trial_choice": "no evaluated run has solved this task yet; this is the failed attempt with the most progress (Codex + GPT-6 Astra (high), privileged, progress 0.96); not a solved run", + "renderer": "isaac-rtx PathTracing spp=32 totalSpp=32 maxBounces=6 optix-denoiser firefly-filter (Isaac Sim 5.1 / Isaac Lab 2.3.2; dedicated render camera at the scene's own viewpoint, rendered natively at 2560x1440, focal 18.0 mm; recorded states posed, render only)", + "rendering": "path-tracing", + "width": 2560, + "height": 1440, + "fps": 30, + "frames": 939, + "seconds": 31.3, + "speed": 16.0, + "capture": "one recorded state every 8 control steps (0.533 s of simulated time) per video frame = 16x real time, 879 rendered frames; 0.5 s hold of the first frame, 1.5 s hold of the final frame", + "camera": { + "eye": [ + -0.35, + -1.9, + 1.75 + ], + "target": [ + -0.4, + 0.0, + 1.05 + ], + "focal_mm": 18.0, + "room_camera": [ + [ + -0.35, + -1.9, + 1.75 + ], + [ + -0.4, + 0.0, + 1.05 + ] + ] + }, + "success_check": false, + "success_check_how": "the video poses the graded episode's own recorded scene states (trajectory.npz frame_state: every asset's root pose and velocity and joint state, recorded during the graded verifier rollout) on the task's own scene build (InteractiveScene.reset_to, no physics step), so it shows the graded run itself. Start state = the frozen instance (instance_ok 1, NVIDIA L40S). The task's success predicate on the posed final state: 0. Verdict (reward.json): success 0, progress 0.9586. The verifier's own camera replay of the actions ended on the graded final state: camera_replay_match 0.", + "image": "efe-private:embodiedswe-0.3.0 (verifier tooling: efe:embodiedswe-files-0.3.1)", + "core": true, + "encode": "libx264 preset slow CRF 16, yuv420p, bt709, +faststart", + "poster": "the final frame", + "created": "2026-10-30", + "why_no_video": "No evaluated run has solved this task yet. The video shows the failed attempt with the most progress (Codex + GPT-6 Astra (high), privileged, progress 0.96), not a solved run.", + "shows": "agent-attempt (failed):codex-gpt6_astra-high/464xuPn" +} \ No newline at end of file diff --git a/masters/embodiedswe/assembly_ikea_table_bimanual_franka_osc/poster.png b/masters/embodiedswe/assembly_ikea_table_bimanual_franka_osc/poster.png new file mode 100644 index 0000000000000000000000000000000000000000..648605dc580ee4b4a125792acc42b6dd00ec01b0 --- /dev/null +++ b/masters/embodiedswe/assembly_ikea_table_bimanual_franka_osc/poster.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9f71cbfa1f95b812f534f150689a936ad55d3586c67214c3691941e41a76ef55 +size 2790999 diff --git a/masters/embodiedswe/assembly_ikea_table_bimanual_franka_osc/video.mp4 b/masters/embodiedswe/assembly_ikea_table_bimanual_franka_osc/video.mp4 new file mode 100644 index 0000000000000000000000000000000000000000..a70118717a833694c59704b0030004c47d231639 --- /dev/null +++ b/masters/embodiedswe/assembly_ikea_table_bimanual_franka_osc/video.mp4 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b0fb71cc5a0fef286076bc6477046b431a7788406034e0d45efc1c1651116bcc +size 14786984 diff --git a/masters/embodiedswe/assembly_nut_thread_franka_osc/meta.json b/masters/embodiedswe/assembly_nut_thread_franka_osc/meta.json new file mode 100644 index 0000000000000000000000000000000000000000..5e888a1209b5ab9fbd3a832b37e19eb15ed5a0f1 --- /dev/null +++ b/masters/embodiedswe/assembly_nut_thread_franka_osc/meta.json @@ -0,0 +1,52 @@ +{ + "suite": "embodiedswe", + "task": "assembly_nut_thread_franka_osc", + "task_dir": "embodiedswe-assembly-nut-thread-franka-osc-i00-standard", + "source": "agent-success:codex-gpt6_luna-xhigh-embodiedswe-assembly-nut-thread-franka-osc-i00-standard/embodiedswe-assembly-nut-thread__Vw6ueVC", + "agent": "Codex + GPT-6 Luna (xhigh)", + "run": "codex-gpt6_luna-xhigh", + "mode": "standard", + "trial": "Vw6ueVC", + "trial_choice": "the only run that solved this task (GPT-6 Luna, counted standard success); the other runs did not attempt it (screening: they run only tasks Luna fails in every mode)", + "renderer": "isaac-rtx PathTracing spp=32 totalSpp=32 maxBounces=6 optix-denoiser firefly-filter (Isaac Sim 5.1 / Isaac Lab 2.3.2; dedicated render camera at the scene's own viewpoint, rendered natively at 2560x1440, focal 18.0 mm; recorded states posed, render only)", + "rendering": "path-tracing", + "width": 2560, + "height": 1440, + "fps": 30, + "frames": 834, + "seconds": 27.8, + "speed": 16.0, + "capture": "one recorded state every 8 control steps (0.533 s of simulated time) per video frame = 16x real time, 774 rendered frames; 0.5 s hold of the first frame, 1.5 s hold of the final frame", + "camera": { + "eye": [ + 1.1, + -0.72, + 0.85 + ], + "target": [ + 0.36, + 0.05, + 0.2 + ], + "focal_mm": 18.0, + "room_camera": [ + [ + 1.1, + -0.72, + 0.85 + ], + [ + 0.36, + 0.05, + 0.2 + ] + ] + }, + "success_check": true, + "success_check_how": "the video poses the graded episode's own recorded scene states (trajectory.npz frame_state: every asset's root pose and velocity and joint state, recorded during the graded live standard episode by the simulator service) on the task's own scene build (InteractiveScene.reset_to, no physics step), so it shows the graded run itself. Start state = the frozen instance (instance_ok 1, NVIDIA L40S). The task's success predicate on the posed final state: 1. Verdict (reward.json): success 1, progress 1.0.", + "image": "efe-private:embodiedswe-0.3.0 (sim service: efe:embodiedswe-standard-files-0.2.0)", + "core": false, + "encode": "libx264 preset slow CRF 16, yuv420p, bt709, +faststart", + "poster": "the final frame", + "created": "2026-10-30" +} \ No newline at end of file diff --git a/masters/embodiedswe/assembly_nut_thread_franka_osc/poster.png b/masters/embodiedswe/assembly_nut_thread_franka_osc/poster.png new file mode 100644 index 0000000000000000000000000000000000000000..9f0c9d7a6cd9ae3c6c010c6cdf6121a21d0cda6c --- /dev/null +++ b/masters/embodiedswe/assembly_nut_thread_franka_osc/poster.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:84c4917cde65805895d89159e277d60e96528dd5dfb2552b853bce5babc4fe68 +size 2866443 diff --git a/masters/embodiedswe/assembly_nut_thread_franka_osc/video.mp4 b/masters/embodiedswe/assembly_nut_thread_franka_osc/video.mp4 new file mode 100644 index 0000000000000000000000000000000000000000..2cd81717ced73f29ffe5f845edf8648e6b093374 --- /dev/null +++ b/masters/embodiedswe/assembly_nut_thread_franka_osc/video.mp4 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:099a43130b8b47c98b050ba2dd3292692776905e1e473a451e5d00aee851b931 +size 17001493 diff --git a/masters/embodiedswe/assembly_pc_gpu_franka_osc/meta.json b/masters/embodiedswe/assembly_pc_gpu_franka_osc/meta.json new file mode 100644 index 0000000000000000000000000000000000000000..4e198aa8f7195f0144afc52ddeb509d23e471bc0 --- /dev/null +++ b/masters/embodiedswe/assembly_pc_gpu_franka_osc/meta.json @@ -0,0 +1,52 @@ +{ + "suite": "embodiedswe", + "task": "assembly_pc_gpu_franka_osc", + "task_dir": "embodiedswe-assembly-pc-gpu-franka-osc-i00-standard", + "source": "agent-success:codex-gpt6_astra-high-embodiedswe-assembly-pc-gpu-franka-osc-i00-standard/embodiedswe-assembly-pc-gpu-fran__AoPoRzm", + "agent": "Codex + GPT-6 Astra (high)", + "run": "codex-gpt6_astra-high", + "mode": "standard", + "trial": "AoPoRzm", + "trial_choice": "counted standard success; order: GPT-6 Astra > Claude Opus 5.5 > Gemini 4 Argon > Gemini 3.8 Flash, then the task's default mode (standard) before privileged", + "renderer": "isaac-rtx PathTracing spp=32 totalSpp=32 maxBounces=6 optix-denoiser firefly-filter (Isaac Sim 5.1 / Isaac Lab 2.3.2; dedicated render camera at the scene's own viewpoint, rendered natively at 2560x1440, focal 18.0 mm; recorded states posed, render only)", + "rendering": "path-tracing", + "width": 2560, + "height": 1440, + "fps": 30, + "frames": 693, + "seconds": 23.1, + "speed": 24.0, + "capture": "one recorded state every 12 control steps (0.800 s of simulated time) per video frame = 24x real time, 633 rendered frames; 0.5 s hold of the first frame, 1.5 s hold of the final frame", + "camera": { + "eye": [ + 1.22, + -1.2, + 1.08 + ], + "target": [ + 0.3, + -0.02, + 0.2 + ], + "focal_mm": 18.0, + "room_camera": [ + [ + 1.22, + -1.2, + 1.08 + ], + [ + 0.3, + -0.02, + 0.2 + ] + ] + }, + "success_check": true, + "success_check_how": "the video poses the graded episode's own recorded scene states (trajectory.npz frame_state: every asset's root pose and velocity and joint state, recorded during the graded live standard episode by the simulator service) on the task's own scene build (InteractiveScene.reset_to, no physics step), so it shows the graded run itself. Start state = the frozen instance (instance_ok 1, NVIDIA L40S). The task's success predicate on the posed final state: 1. Verdict (reward.json): success 1, progress 1.0.", + "image": "efe-private:embodiedswe-0.3.0 (sim service: efe:embodiedswe-standard-files-0.2.0)", + "core": true, + "encode": "libx264 preset slow CRF 16, yuv420p, bt709, +faststart", + "poster": "the final frame", + "created": "2026-10-30" +} \ No newline at end of file diff --git a/masters/embodiedswe/assembly_pc_gpu_franka_osc/poster.png b/masters/embodiedswe/assembly_pc_gpu_franka_osc/poster.png new file mode 100644 index 0000000000000000000000000000000000000000..5f99c88c9fd5681d5e4daf7610b2509597fcd8cc --- /dev/null +++ b/masters/embodiedswe/assembly_pc_gpu_franka_osc/poster.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7a0bfd255229e945330dbf9abd1fab765546a998b526eacc8386e79ab7b83f26 +size 2686747 diff --git a/masters/embodiedswe/assembly_pc_gpu_franka_osc/video.mp4 b/masters/embodiedswe/assembly_pc_gpu_franka_osc/video.mp4 new file mode 100644 index 0000000000000000000000000000000000000000..47eca9a44ed076517990580ef28734f9a451bacb --- /dev/null +++ b/masters/embodiedswe/assembly_pc_gpu_franka_osc/video.mp4 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7cd461aacc297eff983f8727b9096fbacdbf435ad0c4af8617ca1fd9cdd83514 +size 11711462 diff --git a/masters/embodiedswe/assembly_pc_gpu_ram_franka_osc/meta.json b/masters/embodiedswe/assembly_pc_gpu_ram_franka_osc/meta.json new file mode 100644 index 0000000000000000000000000000000000000000..eecd668739c02aa4a27381f869d7b1b06d18963e --- /dev/null +++ b/masters/embodiedswe/assembly_pc_gpu_ram_franka_osc/meta.json @@ -0,0 +1,52 @@ +{ + "suite": "embodiedswe", + "task": "assembly_pc_gpu_ram_franka_osc", + "task_dir": "embodiedswe-assembly-pc-gpu-ram-franka-osc-i00-privileged", + "source": "agent-success:codex-gpt6_astra-high-embodiedswe-assembly-pc-gpu-ram-franka-osc-i00-privileged/embodiedswe-assembly-pc-gpu-ram__7z5tXVi", + "agent": "Codex + GPT-6 Astra (high)", + "run": "codex-gpt6_astra-high", + "mode": "privileged", + "trial": "7z5tXVi", + "trial_choice": "counted privileged success; order: GPT-6 Astra > Claude Opus 5.5 > Gemini 4 Argon > Gemini 3.8 Flash, then the task's default mode (standard) before privileged", + "renderer": "isaac-rtx PathTracing spp=32 totalSpp=32 maxBounces=6 optix-denoiser firefly-filter (Isaac Sim 5.1 / Isaac Lab 2.3.2; dedicated render camera at the scene's own viewpoint, rendered natively at 2560x1440, focal 18.0 mm; recorded states posed, render only)", + "rendering": "path-tracing", + "width": 2560, + "height": 1440, + "fps": 30, + "frames": 672, + "seconds": 22.4, + "speed": 8.0, + "capture": "one recorded state every 4 control steps (0.267 s of simulated time) per video frame = 8x real time, 612 rendered frames; 0.5 s hold of the first frame, 1.5 s hold of the final frame", + "camera": { + "eye": [ + 1.22, + -1.2, + 1.08 + ], + "target": [ + 0.3, + -0.02, + 0.2 + ], + "focal_mm": 18.0, + "room_camera": [ + [ + 1.22, + -1.2, + 1.08 + ], + [ + 0.3, + -0.02, + 0.2 + ] + ] + }, + "success_check": true, + "success_check_how": "the video poses the graded episode's own recorded scene states (trajectory.npz frame_state: every asset's root pose and velocity and joint state, recorded during the graded verifier rollout) on the task's own scene build (InteractiveScene.reset_to, no physics step), so it shows the graded run itself. Start state = the frozen instance (instance_ok 1, NVIDIA L40S). The task's success predicate on the posed final state: 1. Verdict (reward.json): success 1, progress 1.0. The verifier's own camera replay of the actions ended on the graded final state: camera_replay_match 1.", + "image": "efe-private:embodiedswe-0.3.0 (verifier tooling: efe:embodiedswe-files-0.3.1)", + "core": true, + "encode": "libx264 preset slow CRF 16, yuv420p, bt709, +faststart", + "poster": "the final frame", + "created": "2026-10-30" +} \ No newline at end of file diff --git a/masters/embodiedswe/assembly_pc_gpu_ram_franka_osc/poster.png b/masters/embodiedswe/assembly_pc_gpu_ram_franka_osc/poster.png new file mode 100644 index 0000000000000000000000000000000000000000..651abf0e01bd980a0a44bae868546597874d3948 --- /dev/null +++ b/masters/embodiedswe/assembly_pc_gpu_ram_franka_osc/poster.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5432578b12b6547c7a9e1ec5a9c4e50d96e11638c578c5fbffa25f6fcb18e8e5 +size 2681787 diff --git a/masters/embodiedswe/assembly_pc_gpu_ram_franka_osc/video.mp4 b/masters/embodiedswe/assembly_pc_gpu_ram_franka_osc/video.mp4 new file mode 100644 index 0000000000000000000000000000000000000000..8741f8c86d210d283d3fa3ca8d653ccac45d1978 --- /dev/null +++ b/masters/embodiedswe/assembly_pc_gpu_ram_franka_osc/video.mp4 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cd2fbdcf181cedf4eb8c61514e9d8ecdc31fcc337ef552ed3bc7936486d34307 +size 8701633 diff --git a/masters/embodiedswe/assembly_pc_motherboard_franka_osc/meta.json b/masters/embodiedswe/assembly_pc_motherboard_franka_osc/meta.json new file mode 100644 index 0000000000000000000000000000000000000000..8212577680ca2b591864abb59890332b7ab73472 --- /dev/null +++ b/masters/embodiedswe/assembly_pc_motherboard_franka_osc/meta.json @@ -0,0 +1,52 @@ +{ + "suite": "embodiedswe", + "task": "assembly_pc_motherboard_franka_osc", + "task_dir": "embodiedswe-assembly-pc-motherboard-franka-osc-i00-standard", + "source": "agent-success:codex-gpt6_astra-high-embodiedswe-assembly-pc-motherboard-franka-osc-i00-standard/embodiedswe-assembly-pc-motherbo__advYJ7n", + "agent": "Codex + GPT-6 Astra (high)", + "run": "codex-gpt6_astra-high", + "mode": "standard", + "trial": "advYJ7n", + "trial_choice": "counted standard success; order: GPT-6 Astra > Claude Opus 5.5 > Gemini 4 Argon > Gemini 3.8 Flash, then the task's default mode (standard) before privileged", + "renderer": "isaac-rtx PathTracing spp=32 totalSpp=32 maxBounces=6 optix-denoiser firefly-filter (Isaac Sim 5.1 / Isaac Lab 2.3.2; dedicated render camera at the scene's own viewpoint, rendered natively at 2560x1440, focal 18.0 mm; recorded states posed, render only)", + "rendering": "path-tracing", + "width": 2560, + "height": 1440, + "fps": 30, + "frames": 793, + "seconds": 26.43, + "speed": 48.0, + "capture": "one recorded state every 24 control steps (1.600 s of simulated time) per video frame = 48x real time, 733 rendered frames; 0.5 s hold of the first frame, 1.5 s hold of the final frame", + "camera": { + "eye": [ + 1.22, + -1.2, + 1.08 + ], + "target": [ + 0.3, + -0.02, + 0.2 + ], + "focal_mm": 18.0, + "room_camera": [ + [ + 1.22, + -1.2, + 1.08 + ], + [ + 0.3, + -0.02, + 0.2 + ] + ] + }, + "success_check": true, + "success_check_how": "the video poses the graded episode's own recorded scene states (trajectory.npz frame_state: every asset's root pose and velocity and joint state, recorded during the graded live standard episode by the simulator service) on the task's own scene build (InteractiveScene.reset_to, no physics step), so it shows the graded run itself. Start state = the frozen instance (instance_ok 1, NVIDIA L40S). The task's success predicate on the posed final state: 1. Verdict (reward.json): success 1, progress 1.0.", + "image": "efe-private:embodiedswe-0.3.0 (sim service: efe:embodiedswe-standard-files-0.2.0)", + "core": true, + "encode": "libx264 preset slow CRF 16, yuv420p, bt709, +faststart", + "poster": "the final frame", + "created": "2026-10-30" +} \ No newline at end of file diff --git a/masters/embodiedswe/assembly_pc_motherboard_franka_osc/poster.png b/masters/embodiedswe/assembly_pc_motherboard_franka_osc/poster.png new file mode 100644 index 0000000000000000000000000000000000000000..c609b46ef379669740432d6ed038b6ec7f2652c0 --- /dev/null +++ b/masters/embodiedswe/assembly_pc_motherboard_franka_osc/poster.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e03e1ef2981f515bedaad2fbbf8ae59c6c1551e5cea35a3b8f776c1594be4a7e +size 2681928 diff --git a/masters/embodiedswe/assembly_pc_motherboard_franka_osc/video.mp4 b/masters/embodiedswe/assembly_pc_motherboard_franka_osc/video.mp4 new file mode 100644 index 0000000000000000000000000000000000000000..5346eca476a04bdfb1ed1034f2b1b5747174f6fd --- /dev/null +++ b/masters/embodiedswe/assembly_pc_motherboard_franka_osc/video.mp4 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8477997aa6431a7f4f8a6aabaf043a5a606b9b3d8be40f3eb00e9a60688434e3 +size 13527641 diff --git a/masters/embodiedswe/assembly_pc_ram_franka_osc/meta.json b/masters/embodiedswe/assembly_pc_ram_franka_osc/meta.json new file mode 100644 index 0000000000000000000000000000000000000000..c58ee61349f96204ab7912c879799277be8daf45 --- /dev/null +++ b/masters/embodiedswe/assembly_pc_ram_franka_osc/meta.json @@ -0,0 +1,52 @@ +{ + "suite": "embodiedswe", + "task": "assembly_pc_ram_franka_osc", + "task_dir": "embodiedswe-assembly-pc-ram-franka-osc-i00-privileged", + "source": "agent-success:codex-gpt6_luna-xhigh-embodiedswe-assembly-pc-ram-franka-osc-i00-privileged/embodiedswe-assembly-pc-ram-fran__3KkLgig", + "agent": "Codex + GPT-6 Luna (xhigh)", + "run": "codex-gpt6_luna-xhigh", + "mode": "privileged", + "trial": "3KkLgig", + "trial_choice": "the only run that solved this task (GPT-6 Luna, counted privileged success); the other runs did not attempt it (screening: they run only tasks Luna fails in every mode)", + "renderer": "isaac-rtx PathTracing spp=32 totalSpp=32 maxBounces=6 optix-denoiser firefly-filter (Isaac Sim 5.1 / Isaac Lab 2.3.2; dedicated render camera at the scene's own viewpoint, rendered natively at 2560x1440, focal 18.0 mm; recorded states posed, render only)", + "rendering": "path-tracing", + "width": 2560, + "height": 1440, + "fps": 30, + "frames": 703, + "seconds": 23.43, + "speed": 4.0, + "capture": "one recorded state every 2 control steps (0.133 s of simulated time) per video frame = 4x real time, 643 rendered frames; 0.5 s hold of the first frame, 1.5 s hold of the final frame", + "camera": { + "eye": [ + 1.22, + -1.2, + 1.08 + ], + "target": [ + 0.3, + -0.02, + 0.2 + ], + "focal_mm": 18.0, + "room_camera": [ + [ + 1.22, + -1.2, + 1.08 + ], + [ + 0.3, + -0.02, + 0.2 + ] + ] + }, + "success_check": true, + "success_check_how": "the video poses the graded episode's own recorded scene states (trajectory.npz frame_state: every asset's root pose and velocity and joint state, recorded during the graded verifier rollout) on the task's own scene build (InteractiveScene.reset_to, no physics step), so it shows the graded run itself. Start state = the frozen instance (instance_ok 1, NVIDIA L40S). The task's success predicate on the posed final state: 1. Verdict (reward.json): success 1, progress 1.0. The verifier's own camera replay of the actions ended on the graded final state: camera_replay_match 0.", + "image": "efe-private:embodiedswe-0.3.0 (verifier tooling: efe:embodiedswe-files-0.3.1)", + "core": false, + "encode": "libx264 preset slow CRF 16, yuv420p, bt709, +faststart", + "poster": "the final frame", + "created": "2026-10-30" +} \ No newline at end of file diff --git a/masters/embodiedswe/assembly_pc_ram_franka_osc/poster.png b/masters/embodiedswe/assembly_pc_ram_franka_osc/poster.png new file mode 100644 index 0000000000000000000000000000000000000000..4c1b537a51c0d073947ee8d6075430f336c43463 --- /dev/null +++ b/masters/embodiedswe/assembly_pc_ram_franka_osc/poster.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e12b41fa32f86ddaaaab48a8a71b04f4e0977e8c7a62abaf8e6f4cd7a0cbab72 +size 2656051 diff --git a/masters/embodiedswe/assembly_pc_ram_franka_osc/video.mp4 b/masters/embodiedswe/assembly_pc_ram_franka_osc/video.mp4 new file mode 100644 index 0000000000000000000000000000000000000000..6b64f01cd4c3f591d83f73d6237f31d8c2bb768e --- /dev/null +++ b/masters/embodiedswe/assembly_pc_ram_franka_osc/video.mp4 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9739e8da18d00daa7b0bd0d406ad2929db850c0bc2889add75fc909baa494d47 +size 7974405 diff --git a/masters/embodiedswe/assembly_so101_bimanual_franka_osc/meta.json b/masters/embodiedswe/assembly_so101_bimanual_franka_osc/meta.json new file mode 100644 index 0000000000000000000000000000000000000000..50d486565377caebe1f6d94a6b50a70f5c3eac37 --- /dev/null +++ b/masters/embodiedswe/assembly_so101_bimanual_franka_osc/meta.json @@ -0,0 +1,54 @@ +{ + "suite": "embodiedswe", + "task": "assembly_so101_bimanual_franka_osc", + "task_dir": "embodiedswe-assembly-so101-bimanual-franka-osc-i00-privileged", + "source": "scene-orbit", + "agent": "Codex + GPT-6 Astra (high)", + "run": "codex-gpt6_astra-high", + "mode": "privileged", + "trial": "tzC9EQD", + "trial_choice": "no evaluated run has solved this task yet; this is the failed attempt with the most progress (Codex + GPT-6 Astra (high), privileged, progress 0.14); not a solved run", + "renderer": "isaac-rtx PathTracing spp=32 totalSpp=32 maxBounces=6 optix-denoiser firefly-filter (Isaac Sim 5.1 / Isaac Lab 2.3.2; dedicated render camera at the scene's own viewpoint, rendered natively at 2560x1440, focal 18.0 mm; recorded states posed, render only)", + "rendering": "path-tracing", + "width": 2560, + "height": 1440, + "fps": 30, + "frames": 482, + "seconds": 16.07, + "speed": 4.0, + "capture": "one recorded state every 2 control steps (0.133 s of simulated time) per video frame = 4x real time, 422 rendered frames; 0.5 s hold of the first frame, 1.5 s hold of the final frame", + "camera": { + "eye": [ + 1.15, + 1.1, + 1.65 + ], + "target": [ + 0.4, + -0.1, + 1.0 + ], + "focal_mm": 18.0, + "room_camera": [ + [ + 1.15, + 1.1, + 1.65 + ], + [ + 0.4, + -0.1, + 1.0 + ] + ] + }, + "success_check": false, + "success_check_how": "the video poses the graded episode's own recorded scene states (trajectory.npz frame_state: every asset's root pose and velocity and joint state, recorded during the graded verifier rollout) on the task's own scene build (InteractiveScene.reset_to, no physics step), so it shows the graded run itself. Start state = the frozen instance (instance_ok 1, NVIDIA L40S). The task's success predicate on the posed final state: 0. Verdict (reward.json): success 0, progress 0.1429. The verifier's own camera replay of the actions ended on the graded final state: camera_replay_match 0.", + "image": "efe-private:embodiedswe-0.3.0 (verifier tooling: efe:embodiedswe-files-0.3.1)", + "core": true, + "encode": "libx264 preset slow CRF 16, yuv420p, bt709, +faststart", + "poster": "the final frame", + "created": "2026-10-30", + "why_no_video": "No evaluated run has solved this task yet. The video shows the failed attempt with the most progress (Codex + GPT-6 Astra (high), privileged, progress 0.14), not a solved run.", + "shows": "agent-attempt (failed):codex-gpt6_astra-high/tzC9EQD" +} \ No newline at end of file diff --git a/masters/embodiedswe/assembly_so101_bimanual_franka_osc/poster.png b/masters/embodiedswe/assembly_so101_bimanual_franka_osc/poster.png new file mode 100644 index 0000000000000000000000000000000000000000..219513ab22124dc6821e6cd420519f4a5392a915 --- /dev/null +++ b/masters/embodiedswe/assembly_so101_bimanual_franka_osc/poster.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bac414c0b7e50cac6d8a53a0947c4a2bd4b6ab1cf7d1aab118479fc63a0ad639 +size 2975021 diff --git a/masters/embodiedswe/assembly_so101_bimanual_franka_osc/video.mp4 b/masters/embodiedswe/assembly_so101_bimanual_franka_osc/video.mp4 new file mode 100644 index 0000000000000000000000000000000000000000..05e34b7d778be5da2504cdc36bd29c172952e725 --- /dev/null +++ b/masters/embodiedswe/assembly_so101_bimanual_franka_osc/video.mp4 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:99f93a7b5e3e4028c9dbc74789772346ef076d507971982e29ce52520e7510b8 +size 9269918 diff --git a/masters/embodiedswe/cutting_dice_franka_osc/meta.json b/masters/embodiedswe/cutting_dice_franka_osc/meta.json new file mode 100644 index 0000000000000000000000000000000000000000..68b17fb9192e659dd3fa794d0cc44fd520964ad6 --- /dev/null +++ b/masters/embodiedswe/cutting_dice_franka_osc/meta.json @@ -0,0 +1,52 @@ +{ + "suite": "embodiedswe", + "task": "cutting_dice_franka_osc", + "task_dir": "embodiedswe-cutting-dice-franka-osc-i00-standard", + "source": "agent-success:codex-gpt6_astra-high-embodiedswe-cutting-dice-franka-osc-i00-standard/embodiedswe-cutting-dice-franka__QCU2Zao", + "agent": "Codex + GPT-6 Astra (high)", + "run": "codex-gpt6_astra-high", + "mode": "standard", + "trial": "QCU2Zao", + "trial_choice": "counted standard success; order: GPT-6 Astra > Claude Opus 5.5 > Gemini 4 Argon > Gemini 3.8 Flash, then the task's default mode (standard) before privileged", + "renderer": "isaac-rtx PathTracing spp=32 totalSpp=32 maxBounces=6 optix-denoiser firefly-filter (Isaac Sim 5.1 / Isaac Lab 2.3.2; dedicated render camera at the scene's own viewpoint, rendered natively at 2560x1440, focal 18.0 mm; recorded states posed, render only)", + "rendering": "path-tracing", + "width": 2560, + "height": 1440, + "fps": 30, + "frames": 775, + "seconds": 25.83, + "speed": 4.0, + "capture": "one recorded state every 2 control steps (0.133 s of simulated time) per video frame = 4x real time, 715 rendered frames; 0.5 s hold of the first frame, 1.5 s hold of the final frame", + "camera": { + "eye": [ + 0.72, + -0.62, + 1.25 + ], + "target": [ + -0.08, + 0.08, + 0.95 + ], + "focal_mm": 18.0, + "room_camera": [ + [ + 0.72, + -0.62, + 1.25 + ], + [ + -0.08, + 0.08, + 0.95 + ] + ] + }, + "success_check": true, + "success_check_how": "the video poses the graded episode's own recorded scene states (trajectory.npz frame_state: every asset's root pose and velocity and joint state, recorded during the graded live standard episode by the simulator service) on the task's own scene build (InteractiveScene.reset_to, no physics step), so it shows the graded run itself. Start state = the frozen instance (instance_ok 1, NVIDIA L40S). The task's success predicate on the posed final state: 0. Verdict (reward.json): success 1, progress 1.0.", + "image": "efe-private:embodiedswe-0.3.0 (sim service: efe:embodiedswe-standard-files-0.2.0)", + "core": true, + "encode": "libx264 preset slow CRF 16, yuv420p, bt709, +faststart", + "poster": "the final frame", + "created": "2026-10-30" +} \ No newline at end of file diff --git a/masters/embodiedswe/cutting_dice_franka_osc/poster.png b/masters/embodiedswe/cutting_dice_franka_osc/poster.png new file mode 100644 index 0000000000000000000000000000000000000000..218bebd3569cd33f2c9fe76b07320cc1ed87a1ac --- /dev/null +++ b/masters/embodiedswe/cutting_dice_franka_osc/poster.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:df5f2f4060d16e1989ce579e2cd92057fca96f118d3237e4a4841643a2c9f039 +size 4550896 diff --git a/masters/embodiedswe/cutting_dice_franka_osc/video.mp4 b/masters/embodiedswe/cutting_dice_franka_osc/video.mp4 new file mode 100644 index 0000000000000000000000000000000000000000..d0cfbdf92567388b77aba642bc6deefc34632065 --- /dev/null +++ b/masters/embodiedswe/cutting_dice_franka_osc/video.mp4 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d1edd2dc418672b512155ca9acc78348efe48a028c4c897086c6638b1934aba8 +size 13708707 diff --git a/masters/embodiedswe/cutting_slice_franka_osc/meta.json b/masters/embodiedswe/cutting_slice_franka_osc/meta.json new file mode 100644 index 0000000000000000000000000000000000000000..c2831b855908a11037f98db84579060c2667f16a --- /dev/null +++ b/masters/embodiedswe/cutting_slice_franka_osc/meta.json @@ -0,0 +1,52 @@ +{ + "suite": "embodiedswe", + "task": "cutting_slice_franka_osc", + "task_dir": "embodiedswe-cutting-slice-franka-osc-i00-privileged", + "source": "agent-success:codex-gpt6_luna-xhigh-embodiedswe-cutting-slice-franka-osc-i00-privileged/embodiedswe-cutting-slice-franka__w9Df8br", + "agent": "Codex + GPT-6 Luna (xhigh)", + "run": "codex-gpt6_luna-xhigh", + "mode": "privileged", + "trial": "w9Df8br", + "trial_choice": "the only run that solved this task (GPT-6 Luna, counted privileged success); the other runs did not attempt it (screening: they run only tasks Luna fails in every mode)", + "renderer": "isaac-rtx PathTracing spp=32 totalSpp=32 maxBounces=6 optix-denoiser firefly-filter (Isaac Sim 5.1 / Isaac Lab 2.3.2; dedicated render camera at the scene's own viewpoint, rendered natively at 2560x1440, focal 18.0 mm; recorded states posed, render only)", + "rendering": "path-tracing", + "width": 2560, + "height": 1440, + "fps": 30, + "frames": 635, + "seconds": 21.17, + "speed": 8.0, + "capture": "one recorded state every 4 control steps (0.267 s of simulated time) per video frame = 8x real time, 575 rendered frames; 0.5 s hold of the first frame, 1.5 s hold of the final frame", + "camera": { + "eye": [ + 0.72, + -0.62, + 1.25 + ], + "target": [ + -0.08, + 0.08, + 0.95 + ], + "focal_mm": 18.0, + "room_camera": [ + [ + 0.72, + -0.62, + 1.25 + ], + [ + -0.08, + 0.08, + 0.95 + ] + ] + }, + "success_check": true, + "success_check_how": "the video poses the graded episode's own recorded scene states (trajectory.npz frame_state: every asset's root pose and velocity and joint state, recorded during the graded verifier rollout) on the task's own scene build (InteractiveScene.reset_to, no physics step), so it shows the graded run itself. Start state = the frozen instance (instance_ok 1, NVIDIA L40S). The task's success predicate on the posed final state: 0. Verdict (reward.json): success 1, progress 1.0. The verifier's own camera replay of the actions ended on the graded final state: camera_replay_match 1.", + "image": "efe-private:embodiedswe-0.3.0 (verifier tooling: efe:embodiedswe-files-0.3.1)", + "core": false, + "encode": "libx264 preset slow CRF 16, yuv420p, bt709, +faststart", + "poster": "the final frame", + "created": "2026-10-30" +} \ No newline at end of file diff --git a/masters/embodiedswe/cutting_slice_franka_osc/poster.png b/masters/embodiedswe/cutting_slice_franka_osc/poster.png new file mode 100644 index 0000000000000000000000000000000000000000..a778c61595848bfa98766d987583d988e05c7543 --- /dev/null +++ b/masters/embodiedswe/cutting_slice_franka_osc/poster.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0b5864a92a81e7da8cf9322ba780d99944c8a6e68d329e4464955aadf1ca209d +size 4458089 diff --git a/masters/embodiedswe/cutting_slice_franka_osc/video.mp4 b/masters/embodiedswe/cutting_slice_franka_osc/video.mp4 new file mode 100644 index 0000000000000000000000000000000000000000..c1548970d32a0c56277960be06c2efa54c0ccadf --- /dev/null +++ b/masters/embodiedswe/cutting_slice_franka_osc/video.mp4 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f30cf9bb2511f6cb686078644656e9b9a0e3d956f269385e5e9137251f2abcbe +size 13124712 diff --git a/masters/embodiedswe/highlights.json b/masters/embodiedswe/highlights.json new file mode 100644 index 0000000000000000000000000000000000000000..1c2a2eaf12d284ada7a7071d440d1c71fec24db9 --- /dev/null +++ b/masters/embodiedswe/highlights.json @@ -0,0 +1,12 @@ +{"teaser": "puzzle_coffee_franka_osc", + "highlights": ["puzzle_coffee_franka_osc", "puzzle_syringe_bimanual_franka_osc", "assembly_pc_motherboard_franka_osc", + "puzzle_spatula_franka_osc", "packing_egg_carton_g1_pink_ik", "packing_tool_packing_franka_osc", + "puzzle_stack_blocks_g1_pink_ik"], + "why": {"puzzle_coffee_franka_osc": "a full capsule-coffee service in one program: open the bay, load the capsule, close it, stage the mug, start the brew, wait it out and serve the mug; GPT-6 Astra's counted success", + "puzzle_syringe_bimanual_franka_osc": "two arms draw liquid into a syringe and dose it; GPT-6 Astra's counted success", + "assembly_pc_motherboard_franka_osc": "seven bolts driven with an allen key inside a PC case, a long-horizon precision task", + "puzzle_spatula_franka_osc": "a tool-only rule: the bread is flipped and served with the spatula, never touched by the hand", + "packing_egg_carton_g1_pink_ik": "a humanoid places three eggs in a carton and closes the lid", + "packing_tool_packing_franka_osc": "three tools into three drawers: open, place, close, in sequence", + "puzzle_stack_blocks_g1_pink_ik": "a humanoid stacks blocks with its own hands"}, + "why_fewer": ""} diff --git a/masters/embodiedswe/locomanip_box_to_bin_g1_loco_pink_ik/meta.json b/masters/embodiedswe/locomanip_box_to_bin_g1_loco_pink_ik/meta.json new file mode 100644 index 0000000000000000000000000000000000000000..593c8da76e165897c892c9f613e6ee2c1f10fd06 --- /dev/null +++ b/masters/embodiedswe/locomanip_box_to_bin_g1_loco_pink_ik/meta.json @@ -0,0 +1,52 @@ +{ + "suite": "embodiedswe", + "task": "locomanip_box_to_bin_g1_loco_pink_ik", + "task_dir": "embodiedswe-locomanip-box-to-bin-g1-loco-pink-ik-i00-privileged", + "source": "agent-success:codex-gpt6_luna-xhigh-embodiedswe-locomanip-box-to-bin-g1-loco-pink-ik-i00-privileged/embodiedswe-locomanip-box-to-bin__7R4exsh", + "agent": "Codex + GPT-6 Luna (xhigh)", + "run": "codex-gpt6_luna-xhigh", + "mode": "privileged", + "trial": "7R4exsh", + "trial_choice": "the only run that solved this task (GPT-6 Luna, counted privileged success); the other runs did not attempt it (screening: they run only tasks Luna fails in every mode)", + "renderer": "isaac-rtx PathTracing spp=32 totalSpp=32 maxBounces=6 optix-denoiser firefly-filter (Isaac Sim 5.1 / Isaac Lab 2.3.2; dedicated render camera at the scene's own viewpoint, rendered natively at 2560x1440, focal 18.0 mm; recorded states posed, render only)", + "rendering": "path-tracing", + "width": 2560, + "height": 1440, + "fps": 30, + "frames": 755, + "seconds": 25.17, + "speed": 6.0, + "capture": "one recorded state every 10 control steps (0.200 s of simulated time) per video frame = 6x real time, 695 rendered frames; 0.5 s hold of the first frame, 1.5 s hold of the final frame", + "camera": { + "eye": [ + 2.9, + -2.7, + 1.9 + ], + "target": [ + 0.95, + 0.2, + 0.65 + ], + "focal_mm": 18.0, + "room_camera": [ + [ + 2.9, + -2.7, + 1.9 + ], + [ + 0.95, + 0.2, + 0.65 + ] + ] + }, + "success_check": true, + "success_check_how": "the video poses the graded episode's own recorded scene states (trajectory.npz frame_state: every asset's root pose and velocity and joint state, recorded during the graded verifier rollout) on the task's own scene build (InteractiveScene.reset_to, no physics step), so it shows the graded run itself. Start state = the frozen instance (instance_ok 1, NVIDIA L40S). The task's success predicate on the posed final state: 0. Verdict (reward.json): success 1, progress 1.0. The verifier's own camera replay of the actions ended on the graded final state: camera_replay_match 0.", + "image": "efe-private:embodiedswe-0.3.0 (verifier tooling: efe:embodiedswe-files-0.3.1)", + "core": false, + "encode": "libx264 preset slow CRF 16, yuv420p, bt709, +faststart", + "poster": "the final frame", + "created": "2026-10-30" +} \ No newline at end of file diff --git a/masters/embodiedswe/locomanip_box_to_bin_g1_loco_pink_ik/poster.png b/masters/embodiedswe/locomanip_box_to_bin_g1_loco_pink_ik/poster.png new file mode 100644 index 0000000000000000000000000000000000000000..2eb376a7cb43d231f317a2120ddbc5c19ffbb948 --- /dev/null +++ b/masters/embodiedswe/locomanip_box_to_bin_g1_loco_pink_ik/poster.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ff8e5c2bf423602a4caf05700f7a534bb0d029d175de9bd5fc9420cc6eddf8cf +size 3041824 diff --git a/masters/embodiedswe/locomanip_box_to_bin_g1_loco_pink_ik/video.mp4 b/masters/embodiedswe/locomanip_box_to_bin_g1_loco_pink_ik/video.mp4 new file mode 100644 index 0000000000000000000000000000000000000000..82596d3a68c5a0e1cdf9924f9e2d453059c9644a --- /dev/null +++ b/masters/embodiedswe/locomanip_box_to_bin_g1_loco_pink_ik/video.mp4 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:821324c4b48e037ab085580df28b4fd28a3ac72ec84c6d44aff5cd4e62d1104f +size 6428106 diff --git a/masters/embodiedswe/locomanip_fruit_delivery_g1_loco_pink_ik/meta.json b/masters/embodiedswe/locomanip_fruit_delivery_g1_loco_pink_ik/meta.json new file mode 100644 index 0000000000000000000000000000000000000000..c86e7e46946519755acb910d1dde207b6560f5b3 --- /dev/null +++ b/masters/embodiedswe/locomanip_fruit_delivery_g1_loco_pink_ik/meta.json @@ -0,0 +1,52 @@ +{ + "suite": "embodiedswe", + "task": "locomanip_fruit_delivery_g1_loco_pink_ik", + "task_dir": "embodiedswe-locomanip-fruit-delivery-g1-loco-pink-ik-i00-privileged", + "source": "agent-success:codex-gpt6_luna-xhigh-embodiedswe-locomanip-fruit-delivery-g1-loco-pink-ik-i00-privileged/embodiedswe-locomanip-fruit-deli__svUzX8e", + "agent": "Codex + GPT-6 Luna (xhigh)", + "run": "codex-gpt6_luna-xhigh", + "mode": "privileged", + "trial": "svUzX8e", + "trial_choice": "the only run that solved this task (GPT-6 Luna, counted privileged success); the other runs did not attempt it (screening: they run only tasks Luna fails in every mode)", + "renderer": "isaac-rtx PathTracing spp=32 totalSpp=32 maxBounces=6 optix-denoiser firefly-filter (Isaac Sim 5.1 / Isaac Lab 2.3.2; dedicated render camera at the scene's own viewpoint, rendered natively at 2560x1440, focal 18.0 mm; recorded states posed, render only)", + "rendering": "path-tracing", + "width": 2560, + "height": 1440, + "fps": 30, + "frames": 869, + "seconds": 28.97, + "speed": 3.0, + "capture": "one recorded state every 6 control steps (0.100 s of simulated time) per video frame = 3x real time, 809 rendered frames; 0.5 s hold of the first frame, 1.5 s hold of the final frame", + "camera": { + "eye": [ + 3.5, + -4.5, + 3.0 + ], + "target": [ + 0.0, + 0.0, + 0.7 + ], + "focal_mm": 18.0, + "room_camera": [ + [ + 3.5, + -4.5, + 3.0 + ], + [ + 0.0, + 0.0, + 0.7 + ] + ] + }, + "success_check": true, + "success_check_how": "the video poses the graded episode's own recorded scene states (trajectory.npz frame_state: every asset's root pose and velocity and joint state, recorded during the graded verifier rollout) on the task's own scene build (InteractiveScene.reset_to, no physics step), so it shows the graded run itself. Start state = the frozen instance (instance_ok 1, NVIDIA L40S). The task's success predicate on the posed final state: 1. Verdict (reward.json): success 1, progress 1.0. The verifier's own camera replay of the actions ended on the graded final state: camera_replay_match 0.", + "image": "efe-private:embodiedswe-0.3.0 (verifier tooling: efe:embodiedswe-files-0.3.1)", + "core": false, + "encode": "libx264 preset slow CRF 16, yuv420p, bt709, +faststart", + "poster": "the final frame", + "created": "2026-10-30" +} \ No newline at end of file diff --git a/masters/embodiedswe/locomanip_fruit_delivery_g1_loco_pink_ik/poster.png b/masters/embodiedswe/locomanip_fruit_delivery_g1_loco_pink_ik/poster.png new file mode 100644 index 0000000000000000000000000000000000000000..6b48b91016bbf2ea1dd7e30d901f37397294e0ac --- /dev/null +++ b/masters/embodiedswe/locomanip_fruit_delivery_g1_loco_pink_ik/poster.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a3953e77d34570039ab9f7ed4324ef830e45b374c1bd25afe5bb36ed6b44a206 +size 2920748 diff --git a/masters/embodiedswe/locomanip_fruit_delivery_g1_loco_pink_ik/video.mp4 b/masters/embodiedswe/locomanip_fruit_delivery_g1_loco_pink_ik/video.mp4 new file mode 100644 index 0000000000000000000000000000000000000000..5e53c471bbefc6fc5da463f97b2aa2c86c5c71d5 --- /dev/null +++ b/masters/embodiedswe/locomanip_fruit_delivery_g1_loco_pink_ik/video.mp4 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d9cd16e047f58ca6056265182ad60d04a23416cfd90ff9d3f1c20bc99444b821 +size 7692998 diff --git a/masters/embodiedswe/locomanip_wheel_carry_g1_loco_pink_ik/meta.json b/masters/embodiedswe/locomanip_wheel_carry_g1_loco_pink_ik/meta.json new file mode 100644 index 0000000000000000000000000000000000000000..93361e91ea37f48d254da9a7877a9f9b9a25046a --- /dev/null +++ b/masters/embodiedswe/locomanip_wheel_carry_g1_loco_pink_ik/meta.json @@ -0,0 +1,52 @@ +{ + "suite": "embodiedswe", + "task": "locomanip_wheel_carry_g1_loco_pink_ik", + "task_dir": "embodiedswe-locomanip-wheel-carry-g1-loco-pink-ik-i00-privileged", + "source": "agent-success:codex-gpt6_luna-xhigh-embodiedswe-locomanip-wheel-carry-g1-loco-pink-ik-i00-privileged/embodiedswe-locomanip-wheel-carr__UcikH72", + "agent": "Codex + GPT-6 Luna (xhigh)", + "run": "codex-gpt6_luna-xhigh", + "mode": "privileged", + "trial": "UcikH72", + "trial_choice": "the only run that solved this task (GPT-6 Luna, counted privileged success); the other runs did not attempt it (screening: they run only tasks Luna fails in every mode)", + "renderer": "isaac-rtx PathTracing spp=32 totalSpp=32 maxBounces=6 optix-denoiser firefly-filter (Isaac Sim 5.1 / Isaac Lab 2.3.2; dedicated render camera at the scene's own viewpoint, rendered natively at 2560x1440, focal 18.0 mm; recorded states posed, render only)", + "rendering": "path-tracing", + "width": 2560, + "height": 1440, + "fps": 30, + "frames": 548, + "seconds": 18.27, + "speed": 3.0, + "capture": "one recorded state every 5 control steps (0.100 s of simulated time) per video frame = 3x real time, 488 rendered frames; 0.5 s hold of the first frame, 1.5 s hold of the final frame", + "camera": { + "eye": [ + -1.5, + -2.5, + 1.8 + ], + "target": [ + 1.5, + 0.3, + 0.6 + ], + "focal_mm": 18.0, + "room_camera": [ + [ + -1.5, + -2.5, + 1.8 + ], + [ + 1.5, + 0.3, + 0.6 + ] + ] + }, + "success_check": true, + "success_check_how": "the video poses the graded episode's own recorded scene states (trajectory.npz frame_state: every asset's root pose and velocity and joint state, recorded during the graded verifier rollout) on the task's own scene build (InteractiveScene.reset_to, no physics step), so it shows the graded run itself. Start state = the frozen instance (instance_ok 1, NVIDIA L40S). The task's success predicate on the posed final state: 0. Verdict (reward.json): success 1, progress 1.0. The verifier's own camera replay of the actions ended on the graded final state: camera_replay_match 0.", + "image": "efe-private:embodiedswe-0.3.0 (verifier tooling: efe:embodiedswe-files-0.3.1)", + "core": false, + "encode": "libx264 preset slow CRF 16, yuv420p, bt709, +faststart", + "poster": "the final frame", + "created": "2026-10-30" +} \ No newline at end of file diff --git a/masters/embodiedswe/locomanip_wheel_carry_g1_loco_pink_ik/poster.png b/masters/embodiedswe/locomanip_wheel_carry_g1_loco_pink_ik/poster.png new file mode 100644 index 0000000000000000000000000000000000000000..cc8fdecbc9be4ad657b312a575cbeea26d22e9bc --- /dev/null +++ b/masters/embodiedswe/locomanip_wheel_carry_g1_loco_pink_ik/poster.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ea42c23a1cd633bd451c564f9640acc928acaeb4a07dcbf39d2c613d372cbefa +size 3124329 diff --git a/masters/embodiedswe/locomanip_wheel_carry_g1_loco_pink_ik/video.mp4 b/masters/embodiedswe/locomanip_wheel_carry_g1_loco_pink_ik/video.mp4 new file mode 100644 index 0000000000000000000000000000000000000000..4ac5894add89660e2e970da879f4de8cda6b9818 --- /dev/null +++ b/masters/embodiedswe/locomanip_wheel_carry_g1_loco_pink_ik/video.mp4 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:159ba7896dd5abe7fd5d5d26ac6f457d3c009b51b74ce3fbed58a968131ea5a6 +size 7244374 diff --git a/masters/embodiedswe/packing_clear_organic_objects_g1_pink_ik/meta.json b/masters/embodiedswe/packing_clear_organic_objects_g1_pink_ik/meta.json new file mode 100644 index 0000000000000000000000000000000000000000..e74e8a75cbc262d6369f2a215d506298e70777f4 --- /dev/null +++ b/masters/embodiedswe/packing_clear_organic_objects_g1_pink_ik/meta.json @@ -0,0 +1,52 @@ +{ + "suite": "embodiedswe", + "task": "packing_clear_organic_objects_g1_pink_ik", + "task_dir": "embodiedswe-packing-clear-organic-objects-g1-pink-ik-i00-privileged", + "source": "agent-success:codex-gpt6_luna-xhigh-embodiedswe-packing-clear-organic-objects-g1-pink-ik-i00-privileged/embodiedswe-packing-clear-organi__X4Gzjd7", + "agent": "Codex + GPT-6 Luna (xhigh)", + "run": "codex-gpt6_luna-xhigh", + "mode": "privileged", + "trial": "X4Gzjd7", + "trial_choice": "the only run that solved this task (GPT-6 Luna, counted privileged success); the other runs did not attempt it (screening: they run only tasks Luna fails in every mode)", + "renderer": "isaac-rtx PathTracing spp=32 totalSpp=32 maxBounces=6 optix-denoiser firefly-filter (Isaac Sim 5.1 / Isaac Lab 2.3.2; dedicated render camera at the scene's own viewpoint, rendered natively at 2560x1440, focal 18.0 mm; recorded states posed, render only)", + "rendering": "path-tracing", + "width": 2560, + "height": 1440, + "fps": 30, + "frames": 763, + "seconds": 25.43, + "speed": 3.0, + "capture": "one recorded state every 6 control steps (0.100 s of simulated time) per video frame = 3x real time, 703 rendered frames; 0.5 s hold of the first frame, 1.5 s hold of the final frame", + "camera": { + "eye": [ + 1.4, + -1.7, + 1.78 + ], + "target": [ + 0.0, + 0.0, + 0.88 + ], + "focal_mm": 18.0, + "room_camera": [ + [ + 1.4, + -1.7, + 1.78 + ], + [ + 0.0, + 0.0, + 0.88 + ] + ] + }, + "success_check": true, + "success_check_how": "the video poses the graded episode's own recorded scene states (trajectory.npz frame_state: every asset's root pose and velocity and joint state, recorded during the graded verifier rollout) on the task's own scene build (InteractiveScene.reset_to, no physics step), so it shows the graded run itself. Start state = the frozen instance (instance_ok 1, NVIDIA L40S). The task's success predicate on the posed final state: 1. Verdict (reward.json): success 1, progress 1.0. The verifier's own camera replay of the actions ended on the graded final state: camera_replay_match 0.", + "image": "efe-private:embodiedswe-0.3.0 (verifier tooling: efe:embodiedswe-files-0.3.1)", + "core": false, + "encode": "libx264 preset slow CRF 16, yuv420p, bt709, +faststart", + "poster": "the final frame", + "created": "2026-10-30" +} \ No newline at end of file diff --git a/masters/embodiedswe/packing_clear_organic_objects_g1_pink_ik/poster.png b/masters/embodiedswe/packing_clear_organic_objects_g1_pink_ik/poster.png new file mode 100644 index 0000000000000000000000000000000000000000..40a8721f2bde20d2366a0a2a93e062958faaf6df --- /dev/null +++ b/masters/embodiedswe/packing_clear_organic_objects_g1_pink_ik/poster.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:99d6b381a492184682a7522bd4ed27b6277438aa79453d455aa32fb69b5bbe92 +size 4230109 diff --git a/masters/embodiedswe/packing_clear_organic_objects_g1_pink_ik/video.mp4 b/masters/embodiedswe/packing_clear_organic_objects_g1_pink_ik/video.mp4 new file mode 100644 index 0000000000000000000000000000000000000000..703926524bad97692935cdec75f2e17be84ebb19 --- /dev/null +++ b/masters/embodiedswe/packing_clear_organic_objects_g1_pink_ik/video.mp4 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:74062a89fa7f11d69c697022dcfb1ef043e3f6f312d1c3f0bdf47cee9e953a28 +size 12232748 diff --git a/masters/embodiedswe/packing_egg_carton_g1_pink_ik/meta.json b/masters/embodiedswe/packing_egg_carton_g1_pink_ik/meta.json new file mode 100644 index 0000000000000000000000000000000000000000..6a5c7c79869570a1e0cfa5ad54b1054f558a5cba --- /dev/null +++ b/masters/embodiedswe/packing_egg_carton_g1_pink_ik/meta.json @@ -0,0 +1,52 @@ +{ + "suite": "embodiedswe", + "task": "packing_egg_carton_g1_pink_ik", + "task_dir": "embodiedswe-packing-egg-carton-g1-pink-ik-i00-privileged", + "source": "agent-success:codex-gpt6_astra-high-embodiedswe-packing-egg-carton-g1-pink-ik-i00-privileged/embodiedswe-packing-egg-carton-g__65i4b65", + "agent": "Codex + GPT-6 Astra (high)", + "run": "codex-gpt6_astra-high", + "mode": "privileged", + "trial": "65i4b65", + "trial_choice": "counted privileged success; order: GPT-6 Astra > Claude Opus 5.5 > Gemini 4 Argon > Gemini 3.8 Flash, then the task's default mode (standard) before privileged", + "renderer": "isaac-rtx PathTracing spp=32 totalSpp=32 maxBounces=6 optix-denoiser firefly-filter (Isaac Sim 5.1 / Isaac Lab 2.3.2; dedicated render camera at the scene's own viewpoint, rendered natively at 2560x1440, focal 18.0 mm; recorded states posed, render only)", + "rendering": "path-tracing", + "width": 2560, + "height": 1440, + "fps": 30, + "frames": 772, + "seconds": 25.73, + "speed": 6.0, + "capture": "one recorded state every 12 control steps (0.200 s of simulated time) per video frame = 6x real time, 712 rendered frames; 0.5 s hold of the first frame, 1.5 s hold of the final frame", + "camera": { + "eye": [ + 1.25, + 0.15, + 1.35 + ], + "target": [ + 0.0, + -0.42, + 0.78 + ], + "focal_mm": 18.0, + "room_camera": [ + [ + 1.25, + 0.15, + 1.3499999999999999 + ], + [ + 0.0, + -0.42, + 0.78 + ] + ] + }, + "success_check": true, + "success_check_how": "the video poses the graded episode's own recorded scene states (trajectory.npz frame_state: every asset's root pose and velocity and joint state, recorded during the graded verifier rollout) on the task's own scene build (InteractiveScene.reset_to, no physics step), so it shows the graded run itself. Start state = the frozen instance (instance_ok 1, NVIDIA L40S). The task's success predicate on the posed final state: 1. Verdict (reward.json): success 1, progress 1.0. The verifier's own camera replay of the actions ended on the graded final state: camera_replay_match 0.", + "image": "efe-private:embodiedswe-0.3.0 (verifier tooling: efe:embodiedswe-files-0.3.1)", + "core": true, + "encode": "libx264 preset slow CRF 16, yuv420p, bt709, +faststart", + "poster": "the final frame", + "created": "2026-10-30" +} \ No newline at end of file diff --git a/masters/embodiedswe/packing_egg_carton_g1_pink_ik/poster.png b/masters/embodiedswe/packing_egg_carton_g1_pink_ik/poster.png new file mode 100644 index 0000000000000000000000000000000000000000..44699deb31ec6aa1d8086770e865577ff9722d41 --- /dev/null +++ b/masters/embodiedswe/packing_egg_carton_g1_pink_ik/poster.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8b265b2ad9667ea4538c69bfe84e33fb256d5ac66ec69d27797da7e28bc75dac +size 3897701 diff --git a/masters/embodiedswe/packing_egg_carton_g1_pink_ik/video.mp4 b/masters/embodiedswe/packing_egg_carton_g1_pink_ik/video.mp4 new file mode 100644 index 0000000000000000000000000000000000000000..3e50a612b14eb487cdf1630ca324c1fd2d029903 --- /dev/null +++ b/masters/embodiedswe/packing_egg_carton_g1_pink_ik/video.mp4 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f0ae1f652b83f910b4db0b1a5133c33983e16c5a6c58c9a54615501c9baf5405 +size 12956346 diff --git a/masters/embodiedswe/packing_pen_holder_franka_osc/meta.json b/masters/embodiedswe/packing_pen_holder_franka_osc/meta.json new file mode 100644 index 0000000000000000000000000000000000000000..2d8c610dbbb7f81d1101fa09407f6fc173a9aee5 --- /dev/null +++ b/masters/embodiedswe/packing_pen_holder_franka_osc/meta.json @@ -0,0 +1,52 @@ +{ + "suite": "embodiedswe", + "task": "packing_pen_holder_franka_osc", + "task_dir": "embodiedswe-packing-pen-holder-franka-osc-i00-standard", + "source": "agent-success:codex-gpt6_luna-xhigh-embodiedswe-packing-pen-holder-franka-osc-i00-standard/embodiedswe-packing-pen-holder-f__YhFgS3L", + "agent": "Codex + GPT-6 Luna (xhigh)", + "run": "codex-gpt6_luna-xhigh", + "mode": "standard", + "trial": "YhFgS3L", + "trial_choice": "the only run that solved this task (GPT-6 Luna, counted standard success); the other runs did not attempt it (screening: they run only tasks Luna fails in every mode)", + "renderer": "isaac-rtx PathTracing spp=32 totalSpp=32 maxBounces=6 optix-denoiser firefly-filter (Isaac Sim 5.1 / Isaac Lab 2.3.2; dedicated render camera at the scene's own viewpoint, rendered natively at 2560x1440, focal 18.0 mm; recorded states posed, render only)", + "rendering": "path-tracing", + "width": 2560, + "height": 1440, + "fps": 30, + "frames": 616, + "seconds": 20.53, + "speed": 4.0, + "capture": "one recorded state every 2 control steps (0.133 s of simulated time) per video frame = 4x real time, 556 rendered frames; 0.5 s hold of the first frame, 1.5 s hold of the final frame", + "camera": { + "eye": [ + -1.2, + -1.05, + 1.2 + ], + "target": [ + 0.1, + 0.05, + 0.75 + ], + "focal_mm": 18.0, + "room_camera": [ + [ + -1.2, + -1.05, + 1.2 + ], + [ + 0.1, + 0.05, + 0.75 + ] + ] + }, + "success_check": true, + "success_check_how": "the video poses the graded episode's own recorded scene states (trajectory.npz frame_state: every asset's root pose and velocity and joint state, recorded during the graded live standard episode by the simulator service) on the task's own scene build (InteractiveScene.reset_to, no physics step), so it shows the graded run itself. Start state = the frozen instance (instance_ok 1, NVIDIA L40S). The task's success predicate on the posed final state: 1. Verdict (reward.json): success 1, progress 1.0.", + "image": "efe-private:embodiedswe-0.3.0 (sim service: efe:embodiedswe-standard-files-0.2.0)", + "core": false, + "encode": "libx264 preset slow CRF 16, yuv420p, bt709, +faststart", + "poster": "the final frame", + "created": "2026-10-30" +} \ No newline at end of file diff --git a/masters/embodiedswe/packing_pen_holder_franka_osc/poster.png b/masters/embodiedswe/packing_pen_holder_franka_osc/poster.png new file mode 100644 index 0000000000000000000000000000000000000000..b3e6cfd4b583447e7b72cad5538ff5097fac9913 --- /dev/null +++ b/masters/embodiedswe/packing_pen_holder_franka_osc/poster.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:05c7d7c5126f8000a1eb49e76525da469e9407e7430fb09a063f8e13e128f5ac +size 3378309 diff --git a/masters/embodiedswe/packing_pen_holder_franka_osc/video.mp4 b/masters/embodiedswe/packing_pen_holder_franka_osc/video.mp4 new file mode 100644 index 0000000000000000000000000000000000000000..08a5c8faa69475b5a7b79cabf8e7bc5e3c2746e5 --- /dev/null +++ b/masters/embodiedswe/packing_pen_holder_franka_osc/video.mp4 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ca7fab250b3bf134ad99b3d2bd7e97a1de2a77dbde5c485d15bb1a25b35e3e6a +size 11260131 diff --git a/masters/embodiedswe/packing_tool_packing_franka_osc/meta.json b/masters/embodiedswe/packing_tool_packing_franka_osc/meta.json new file mode 100644 index 0000000000000000000000000000000000000000..2b530142d12b003e1bfd218c35d7fea9aacc0d3c --- /dev/null +++ b/masters/embodiedswe/packing_tool_packing_franka_osc/meta.json @@ -0,0 +1,52 @@ +{ + "suite": "embodiedswe", + "task": "packing_tool_packing_franka_osc", + "task_dir": "embodiedswe-packing-tool-packing-franka-osc-i00-privileged", + "source": "agent-success:codex-gpt6_astra-high-embodiedswe-packing-tool-packing-franka-osc-i00-privileged/embodiedswe-packing-tool-packing__946VLzS", + "agent": "Codex + GPT-6 Astra (high)", + "run": "codex-gpt6_astra-high", + "mode": "privileged", + "trial": "946VLzS", + "trial_choice": "counted privileged success; order: GPT-6 Astra > Claude Opus 5.5 > Gemini 4 Argon > Gemini 3.8 Flash, then the task's default mode (standard) before privileged", + "renderer": "isaac-rtx PathTracing spp=32 totalSpp=32 maxBounces=6 optix-denoiser firefly-filter (Isaac Sim 5.1 / Isaac Lab 2.3.2; dedicated render camera at the scene's own viewpoint, rendered natively at 2560x1440, focal 18.0 mm; recorded states posed, render only)", + "rendering": "path-tracing", + "width": 2560, + "height": 1440, + "fps": 30, + "frames": 774, + "seconds": 25.8, + "speed": 12.0, + "capture": "one recorded state every 6 control steps (0.400 s of simulated time) per video frame = 12x real time, 714 rendered frames; 0.5 s hold of the first frame, 1.5 s hold of the final frame", + "camera": { + "eye": [ + -1.2, + -1.05, + 1.2 + ], + "target": [ + 0.1, + 0.05, + 0.75 + ], + "focal_mm": 18.0, + "room_camera": [ + [ + -1.2, + -1.05, + 1.2 + ], + [ + 0.1, + 0.05, + 0.75 + ] + ] + }, + "success_check": true, + "success_check_how": "the video poses the graded episode's own recorded scene states (trajectory.npz frame_state: every asset's root pose and velocity and joint state, recorded during the graded verifier rollout) on the task's own scene build (InteractiveScene.reset_to, no physics step), so it shows the graded run itself. Start state = the frozen instance (instance_ok 1, NVIDIA L40S). The task's success predicate on the posed final state: 1. Verdict (reward.json): success 1, progress 1.0. The verifier's own camera replay of the actions ended on the graded final state: camera_replay_match 1.", + "image": "efe-private:embodiedswe-0.3.0 (verifier tooling: efe:embodiedswe-files-0.3.1)", + "core": true, + "encode": "libx264 preset slow CRF 16, yuv420p, bt709, +faststart", + "poster": "the final frame", + "created": "2026-10-30" +} \ No newline at end of file diff --git a/masters/embodiedswe/packing_tool_packing_franka_osc/poster.png b/masters/embodiedswe/packing_tool_packing_franka_osc/poster.png new file mode 100644 index 0000000000000000000000000000000000000000..de2ba1c732cacbab48ce962ef9ca8d4ea6b158b9 --- /dev/null +++ b/masters/embodiedswe/packing_tool_packing_franka_osc/poster.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f0881681be294d1a7a63f53c2e7e4b7c1a827054f7192c8ab04fbed9e47379ce +size 3282451 diff --git a/masters/embodiedswe/packing_tool_packing_franka_osc/video.mp4 b/masters/embodiedswe/packing_tool_packing_franka_osc/video.mp4 new file mode 100644 index 0000000000000000000000000000000000000000..35ff0dee203360d3bb44f518cefb4973db52733d --- /dev/null +++ b/masters/embodiedswe/packing_tool_packing_franka_osc/video.mp4 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ebdd4b2d3feb5c1004ec756477a60221606fd17c67cbaabfb8d589cbe6fa6695 +size 15355742 diff --git a/masters/embodiedswe/puzzle_classify_objects_g1_pink_ik/meta.json b/masters/embodiedswe/puzzle_classify_objects_g1_pink_ik/meta.json new file mode 100644 index 0000000000000000000000000000000000000000..c5c4cc008f95c6f9a4b191ac193712d60767a8a2 --- /dev/null +++ b/masters/embodiedswe/puzzle_classify_objects_g1_pink_ik/meta.json @@ -0,0 +1,52 @@ +{ + "suite": "embodiedswe", + "task": "puzzle_classify_objects_g1_pink_ik", + "task_dir": "embodiedswe-puzzle-classify-objects-g1-pink-ik-i00-standard", + "source": "agent-success:codex-gpt6_luna-xhigh-embodiedswe-puzzle-classify-objects-g1-pink-ik-i00-standard/embodiedswe-puzzle-classify-obje__BCJ4SUj", + "agent": "Codex + GPT-6 Luna (xhigh)", + "run": "codex-gpt6_luna-xhigh", + "mode": "standard", + "trial": "BCJ4SUj", + "trial_choice": "the only run that solved this task (GPT-6 Luna, counted standard success); the other runs did not attempt it (screening: they run only tasks Luna fails in every mode)", + "renderer": "isaac-rtx PathTracing spp=32 totalSpp=32 maxBounces=6 optix-denoiser firefly-filter (Isaac Sim 5.1 / Isaac Lab 2.3.2; dedicated render camera at the scene's own viewpoint, rendered natively at 2560x1440, focal 18.0 mm; recorded states posed, render only)", + "rendering": "path-tracing", + "width": 2560, + "height": 1440, + "fps": 30, + "frames": 324, + "seconds": 10.8, + "speed": 3.0, + "capture": "one recorded state every 6 control steps (0.100 s of simulated time) per video frame = 3x real time, 264 rendered frames; 0.5 s hold of the first frame, 1.5 s hold of the final frame", + "camera": { + "eye": [ + 1.5, + -1.8, + 1.8 + ], + "target": [ + 0.0, + 0.0, + 0.8 + ], + "focal_mm": 18.0, + "room_camera": [ + [ + 1.5, + -1.8, + 1.8 + ], + [ + 0.0, + 0.0, + 0.7999999999999999 + ] + ] + }, + "success_check": true, + "success_check_how": "the video poses the graded episode's own recorded scene states (trajectory.npz frame_state: every asset's root pose and velocity and joint state, recorded during the graded live standard episode by the simulator service) on the task's own scene build (InteractiveScene.reset_to, no physics step), so it shows the graded run itself. Start state = the frozen instance (instance_ok 1, NVIDIA L40S). The task's success predicate on the posed final state: 1. Verdict (reward.json): success 1, progress 1.0.", + "image": "efe-private:embodiedswe-0.3.0 (sim service: efe:embodiedswe-standard-files-0.2.0)", + "core": false, + "encode": "libx264 preset slow CRF 16, yuv420p, bt709, +faststart", + "poster": "the final frame", + "created": "2026-10-30" +} \ No newline at end of file diff --git a/masters/embodiedswe/puzzle_classify_objects_g1_pink_ik/poster.png b/masters/embodiedswe/puzzle_classify_objects_g1_pink_ik/poster.png new file mode 100644 index 0000000000000000000000000000000000000000..311e1cbf8ff281b18cbb69868061e0363078d449 --- /dev/null +++ b/masters/embodiedswe/puzzle_classify_objects_g1_pink_ik/poster.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2ec4edabf7b57c2d0754ed35af69437e8ffd7d36e42ddb8b533f863de3ee91b7 +size 3481429 diff --git a/masters/embodiedswe/puzzle_classify_objects_g1_pink_ik/video.mp4 b/masters/embodiedswe/puzzle_classify_objects_g1_pink_ik/video.mp4 new file mode 100644 index 0000000000000000000000000000000000000000..5013cdad8b65a895beb2dfdf00b74dabd9ebb332 --- /dev/null +++ b/masters/embodiedswe/puzzle_classify_objects_g1_pink_ik/video.mp4 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:67b9ebcd4a65153630ccf970572fadfc8164a42c236c68cb3343bd24b549ff9d +size 4744948 diff --git a/masters/embodiedswe/puzzle_coffee_franka_osc/meta.json b/masters/embodiedswe/puzzle_coffee_franka_osc/meta.json new file mode 100644 index 0000000000000000000000000000000000000000..a89c80492fb19e2baf60f9d798c1bcf7b968479e --- /dev/null +++ b/masters/embodiedswe/puzzle_coffee_franka_osc/meta.json @@ -0,0 +1,52 @@ +{ + "suite": "embodiedswe", + "task": "puzzle_coffee_franka_osc", + "task_dir": "embodiedswe-puzzle-coffee-franka-osc-i00-privileged", + "source": "agent-success:codex-gpt6_astra-high-embodiedswe-puzzle-coffee-franka-osc-i00-privileged/embodiedswe-puzzle-coffee-franka__Xss7c8u", + "agent": "Codex + GPT-6 Astra (high)", + "run": "codex-gpt6_astra-high", + "mode": "privileged", + "trial": "Xss7c8u", + "trial_choice": "counted privileged success; order: GPT-6 Astra > Claude Opus 5.5 > Gemini 4 Argon > Gemini 3.8 Flash, then the task's default mode (standard) before privileged", + "renderer": "isaac-rtx PathTracing spp=32 totalSpp=32 maxBounces=6 optix-denoiser firefly-filter (Isaac Sim 5.1 / Isaac Lab 2.3.2; dedicated render camera at the scene's own viewpoint, rendered natively at 2560x1440, focal 18.0 mm; recorded states posed, render only)", + "rendering": "path-tracing", + "width": 2560, + "height": 1440, + "fps": 30, + "frames": 827, + "seconds": 27.57, + "speed": 12.0, + "capture": "one recorded state every 6 control steps (0.400 s of simulated time) per video frame = 12x real time, 767 rendered frames; 0.5 s hold of the first frame, 1.5 s hold of the final frame", + "camera": { + "eye": [ + 1.5, + -1.7, + 1.55 + ], + "target": [ + 0.0, + 0.0, + 0.65 + ], + "focal_mm": 18.0, + "room_camera": [ + [ + 1.5, + -1.7, + 1.55 + ], + [ + 0.0, + 0.0, + 0.65 + ] + ] + }, + "success_check": true, + "success_check_how": "the video poses the graded episode's own recorded scene states (trajectory.npz frame_state: every asset's root pose and velocity and joint state, recorded during the graded verifier rollout) on the task's own scene build (InteractiveScene.reset_to, no physics step), so it shows the graded run itself. Start state = the frozen instance (instance_ok 1, NVIDIA L40S). The task's success predicate on the posed final state: 0. Verdict (reward.json): success 1, progress 1.0. The verifier's own camera replay of the actions ended on the graded final state: camera_replay_match 1.", + "image": "efe-private:embodiedswe-0.3.0 (verifier tooling: efe:embodiedswe-files-0.3.1)", + "core": true, + "encode": "libx264 preset slow CRF 16, yuv420p, bt709, +faststart", + "poster": "the final frame", + "created": "2026-10-30" +} \ No newline at end of file diff --git a/masters/embodiedswe/puzzle_coffee_franka_osc/poster.png b/masters/embodiedswe/puzzle_coffee_franka_osc/poster.png new file mode 100644 index 0000000000000000000000000000000000000000..38c9f0b76e60107ca857e5b58c70e597c17f57ea --- /dev/null +++ b/masters/embodiedswe/puzzle_coffee_franka_osc/poster.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7029177245f6272ace60bdf39ea0724845649799b7e2dae4c7eabf5e582eda46 +size 3340656 diff --git a/masters/embodiedswe/puzzle_coffee_franka_osc/video.mp4 b/masters/embodiedswe/puzzle_coffee_franka_osc/video.mp4 new file mode 100644 index 0000000000000000000000000000000000000000..6f08c1842bc6b0b65af813a16b1586cb4deb1de7 --- /dev/null +++ b/masters/embodiedswe/puzzle_coffee_franka_osc/video.mp4 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6d83c42fb7c70307dc280fc80b928e6cb8e4801de73021fcb539bcde954b3a10 +size 11038759 diff --git a/masters/embodiedswe/puzzle_push_shapes_g1_pink_ik/meta.json b/masters/embodiedswe/puzzle_push_shapes_g1_pink_ik/meta.json new file mode 100644 index 0000000000000000000000000000000000000000..ff4c2a2b9d2e2da317373443f40e4ec338ec1159 --- /dev/null +++ b/masters/embodiedswe/puzzle_push_shapes_g1_pink_ik/meta.json @@ -0,0 +1,52 @@ +{ + "suite": "embodiedswe", + "task": "puzzle_push_shapes_g1_pink_ik", + "task_dir": "embodiedswe-puzzle-push-shapes-g1-pink-ik-i00-privileged", + "source": "agent-success:codex-gpt6_luna-xhigh-embodiedswe-puzzle-push-shapes-g1-pink-ik-i00-privileged/embodiedswe-puzzle-push-shapes-g__Cfje8BJ", + "agent": "Codex + GPT-6 Luna (xhigh)", + "run": "codex-gpt6_luna-xhigh", + "mode": "privileged", + "trial": "Cfje8BJ", + "trial_choice": "the only run that solved this task (GPT-6 Luna, counted privileged success); the other runs did not attempt it (screening: they run only tasks Luna fails in every mode)", + "renderer": "isaac-rtx PathTracing spp=32 totalSpp=32 maxBounces=6 optix-denoiser firefly-filter (Isaac Sim 5.1 / Isaac Lab 2.3.2; dedicated render camera at the scene's own viewpoint, rendered natively at 2560x1440, focal 18.0 mm; recorded states posed, render only)", + "rendering": "path-tracing", + "width": 2560, + "height": 1440, + "fps": 30, + "frames": 683, + "seconds": 22.77, + "speed": 3.0, + "capture": "one recorded state every 6 control steps (0.100 s of simulated time) per video frame = 3x real time, 623 rendered frames; 0.5 s hold of the first frame, 1.5 s hold of the final frame", + "camera": { + "eye": [ + 1.5, + -1.8, + 1.8 + ], + "target": [ + 0.0, + 0.0, + 0.8 + ], + "focal_mm": 18.0, + "room_camera": [ + [ + 1.5, + -1.8, + 1.8 + ], + [ + 0.0, + 0.0, + 0.7999999999999999 + ] + ] + }, + "success_check": true, + "success_check_how": "the video poses the graded episode's own recorded scene states (trajectory.npz frame_state: every asset's root pose and velocity and joint state, recorded during the graded verifier rollout) on the task's own scene build (InteractiveScene.reset_to, no physics step), so it shows the graded run itself. Start state = the frozen instance (instance_ok 1, NVIDIA L40S). The task's success predicate on the posed final state: 1. Verdict (reward.json): success 1, progress 1.0. The verifier's own camera replay of the actions ended on the graded final state: camera_replay_match 1.", + "image": "efe-private:embodiedswe-0.3.0 (verifier tooling: efe:embodiedswe-files-0.3.1)", + "core": false, + "encode": "libx264 preset slow CRF 16, yuv420p, bt709, +faststart", + "poster": "the final frame", + "created": "2026-10-30" +} \ No newline at end of file diff --git a/masters/embodiedswe/puzzle_push_shapes_g1_pink_ik/poster.png b/masters/embodiedswe/puzzle_push_shapes_g1_pink_ik/poster.png new file mode 100644 index 0000000000000000000000000000000000000000..09b6c43a21836bf834597cacf3ab0e00d230006a --- /dev/null +++ b/masters/embodiedswe/puzzle_push_shapes_g1_pink_ik/poster.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ffbc2de5769f96d0fda260f75c73fb6cc5d53642d888ee6bee9f01a42a3443e3 +size 3505992 diff --git a/masters/embodiedswe/puzzle_push_shapes_g1_pink_ik/video.mp4 b/masters/embodiedswe/puzzle_push_shapes_g1_pink_ik/video.mp4 new file mode 100644 index 0000000000000000000000000000000000000000..df1001c7c2a4c14f9525c2094c945d4f0163887a --- /dev/null +++ b/masters/embodiedswe/puzzle_push_shapes_g1_pink_ik/video.mp4 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:37d1c2c0e4e4e161e7c256efcfd1a5d54e794e2302f561488ee148793cf0d066 +size 8520028 diff --git a/masters/embodiedswe/puzzle_spatula_franka_osc/meta.json b/masters/embodiedswe/puzzle_spatula_franka_osc/meta.json new file mode 100644 index 0000000000000000000000000000000000000000..89e050f5765d07dff7d818d2dae8d7439c856645 --- /dev/null +++ b/masters/embodiedswe/puzzle_spatula_franka_osc/meta.json @@ -0,0 +1,52 @@ +{ + "suite": "embodiedswe", + "task": "puzzle_spatula_franka_osc", + "task_dir": "embodiedswe-puzzle-spatula-franka-osc-i00-privileged", + "source": "agent-success:codex-gpt6_astra-high-embodiedswe-puzzle-spatula-franka-osc-i00-privileged/embodiedswe-puzzle-spatula-frank__WyW3mS5", + "agent": "Codex + GPT-6 Astra (high)", + "run": "codex-gpt6_astra-high", + "mode": "privileged", + "trial": "WyW3mS5", + "trial_choice": "counted privileged success; order: GPT-6 Astra > Claude Opus 5.5 > Gemini 4 Argon > Gemini 3.8 Flash, then the task's default mode (standard) before privileged", + "renderer": "isaac-rtx PathTracing spp=32 totalSpp=32 maxBounces=6 optix-denoiser firefly-filter (Isaac Sim 5.1 / Isaac Lab 2.3.2; dedicated render camera at the scene's own viewpoint, rendered natively at 2560x1440, focal 18.0 mm; recorded states posed, render only)", + "rendering": "path-tracing", + "width": 2560, + "height": 1440, + "fps": 30, + "frames": 697, + "seconds": 23.23, + "speed": 12.0, + "capture": "one recorded state every 6 control steps (0.400 s of simulated time) per video frame = 12x real time, 637 rendered frames; 0.5 s hold of the first frame, 1.5 s hold of the final frame", + "camera": { + "eye": [ + 1.3, + -1.5, + 1.1 + ], + "target": [ + -0.15, + 0.0, + 0.15 + ], + "focal_mm": 18.0, + "room_camera": [ + [ + 1.3, + -1.5, + 1.1 + ], + [ + -0.15, + 0.0, + 0.15 + ] + ] + }, + "success_check": true, + "success_check_how": "the video poses the graded episode's own recorded scene states (trajectory.npz frame_state: every asset's root pose and velocity and joint state, recorded during the graded verifier rollout) on the task's own scene build (InteractiveScene.reset_to, no physics step), so it shows the graded run itself. Start state = the frozen instance (instance_ok 1, NVIDIA L40S). The task's success predicate on the posed final state: 0. Verdict (reward.json): success 1, progress 1.0. The verifier's own camera replay of the actions ended on the graded final state: camera_replay_match 1.", + "image": "efe-private:embodiedswe-0.3.0 (verifier tooling: efe:embodiedswe-files-0.3.1)", + "core": true, + "encode": "libx264 preset slow CRF 16, yuv420p, bt709, +faststart", + "poster": "the final frame", + "created": "2026-10-30" +} \ No newline at end of file diff --git a/masters/embodiedswe/puzzle_spatula_franka_osc/poster.png b/masters/embodiedswe/puzzle_spatula_franka_osc/poster.png new file mode 100644 index 0000000000000000000000000000000000000000..ec144ed3f70ce5a320953885a6938e8c3925ed23 --- /dev/null +++ b/masters/embodiedswe/puzzle_spatula_franka_osc/poster.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:12e3139409ac60311e22dd5acbd5fa1c9b4e20b2f06d27fc7d52c27761ea35fd +size 4331778 diff --git a/masters/embodiedswe/puzzle_spatula_franka_osc/video.mp4 b/masters/embodiedswe/puzzle_spatula_franka_osc/video.mp4 new file mode 100644 index 0000000000000000000000000000000000000000..9fa8c19c9c477951208b6087ddb370b4bef033af --- /dev/null +++ b/masters/embodiedswe/puzzle_spatula_franka_osc/video.mp4 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:63055f927570b70784df0547d97f5ba1492f9999621412250b6839eabf564a6f +size 9265923 diff --git a/masters/embodiedswe/puzzle_stack_blocks_g1_pink_ik/meta.json b/masters/embodiedswe/puzzle_stack_blocks_g1_pink_ik/meta.json new file mode 100644 index 0000000000000000000000000000000000000000..bd4ac6b9cc282640d9576a985472e9812a5de96b --- /dev/null +++ b/masters/embodiedswe/puzzle_stack_blocks_g1_pink_ik/meta.json @@ -0,0 +1,52 @@ +{ + "suite": "embodiedswe", + "task": "puzzle_stack_blocks_g1_pink_ik", + "task_dir": "embodiedswe-puzzle-stack-blocks-g1-pink-ik-i00-standard", + "source": "agent-success:codex-gpt6_astra-high-embodiedswe-puzzle-stack-blocks-g1-pink-ik-i00-standard/embodiedswe-puzzle-stack-blocks__dL6FV5E", + "agent": "Codex + GPT-6 Astra (high)", + "run": "codex-gpt6_astra-high", + "mode": "standard", + "trial": "dL6FV5E", + "trial_choice": "counted standard success; order: GPT-6 Astra > Claude Opus 5.5 > Gemini 4 Argon > Gemini 3.8 Flash, then the task's default mode (standard) before privileged", + "renderer": "isaac-rtx PathTracing spp=32 totalSpp=32 maxBounces=6 optix-denoiser firefly-filter (Isaac Sim 5.1 / Isaac Lab 2.3.2; dedicated render camera at the scene's own viewpoint, rendered natively at 2560x1440, focal 18.0 mm; recorded states posed, render only)", + "rendering": "path-tracing", + "width": 2560, + "height": 1440, + "fps": 30, + "frames": 799, + "seconds": 26.63, + "speed": 24.0, + "capture": "one recorded state every 48 control steps (0.800 s of simulated time) per video frame = 24x real time, 739 rendered frames; 0.5 s hold of the first frame, 1.5 s hold of the final frame", + "camera": { + "eye": [ + 1.5, + -1.8, + 1.8 + ], + "target": [ + 0.0, + 0.0, + 0.8 + ], + "focal_mm": 18.0, + "room_camera": [ + [ + 1.5, + -1.8, + 1.8 + ], + [ + 0.0, + 0.0, + 0.7999999999999999 + ] + ] + }, + "success_check": true, + "success_check_how": "the video poses the graded episode's own recorded scene states (trajectory.npz frame_state: every asset's root pose and velocity and joint state, recorded during the graded live standard episode by the simulator service) on the task's own scene build (InteractiveScene.reset_to, no physics step), so it shows the graded run itself. Start state = the frozen instance (instance_ok 1, NVIDIA L40S). The task's success predicate on the posed final state: 1. Verdict (reward.json): success 1, progress 1.0.", + "image": "efe-private:embodiedswe-0.3.0 (sim service: efe:embodiedswe-standard-files-0.2.0)", + "core": true, + "encode": "libx264 preset slow CRF 16, yuv420p, bt709, +faststart", + "poster": "the final frame", + "created": "2026-10-30" +} \ No newline at end of file diff --git a/masters/embodiedswe/puzzle_stack_blocks_g1_pink_ik/poster.png b/masters/embodiedswe/puzzle_stack_blocks_g1_pink_ik/poster.png new file mode 100644 index 0000000000000000000000000000000000000000..616ff25868aceee82466aacc6490173e1002deff --- /dev/null +++ b/masters/embodiedswe/puzzle_stack_blocks_g1_pink_ik/poster.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1892f44e7de52b9d8ca4ec0239a300e11cc86ccb8bf663376fa251740740d555 +size 3466819 diff --git a/masters/embodiedswe/puzzle_stack_blocks_g1_pink_ik/video.mp4 b/masters/embodiedswe/puzzle_stack_blocks_g1_pink_ik/video.mp4 new file mode 100644 index 0000000000000000000000000000000000000000..afdef24ff7592258cfbd52f17b6e7045ec37c8b2 --- /dev/null +++ b/masters/embodiedswe/puzzle_stack_blocks_g1_pink_ik/video.mp4 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c0bfbaaf876d62f21cb227e625f064019aae00cefc1a93a10d0729318180632a +size 11995833 diff --git a/masters/embodiedswe/puzzle_syringe_bimanual_franka_osc/meta.json b/masters/embodiedswe/puzzle_syringe_bimanual_franka_osc/meta.json new file mode 100644 index 0000000000000000000000000000000000000000..b00d5b7b9153ffa7682dd0eaf746fc7899b5b57c --- /dev/null +++ b/masters/embodiedswe/puzzle_syringe_bimanual_franka_osc/meta.json @@ -0,0 +1,52 @@ +{ + "suite": "embodiedswe", + "task": "puzzle_syringe_bimanual_franka_osc", + "task_dir": "embodiedswe-puzzle-syringe-bimanual-franka-osc-i00-privileged", + "source": "agent-success:codex-gpt6_astra-high-embodiedswe-puzzle-syringe-bimanual-franka-osc-i00-privileged/embodiedswe-puzzle-syringe-biman__dU6guwF", + "agent": "Codex + GPT-6 Astra (high)", + "run": "codex-gpt6_astra-high", + "mode": "privileged", + "trial": "dU6guwF", + "trial_choice": "counted privileged success; order: GPT-6 Astra > Claude Opus 5.5 > Gemini 4 Argon > Gemini 3.8 Flash, then the task's default mode (standard) before privileged", + "renderer": "isaac-rtx PathTracing spp=32 totalSpp=32 maxBounces=6 optix-denoiser firefly-filter (Isaac Sim 5.1 / Isaac Lab 2.3.2; dedicated render camera at the scene's own viewpoint, rendered natively at 2560x1440, focal 18.0 mm; recorded states posed, render only)", + "rendering": "path-tracing", + "width": 2560, + "height": 1440, + "fps": 30, + "frames": 784, + "seconds": 26.13, + "speed": 16.0, + "capture": "one recorded state every 8 control steps (0.533 s of simulated time) per video frame = 16x real time, 724 rendered frames; 0.5 s hold of the first frame, 1.5 s hold of the final frame", + "camera": { + "eye": [ + -1.45, + -0.3, + 1.5 + ], + "target": [ + 0.05, + 0.2, + 1.0 + ], + "focal_mm": 18.0, + "room_camera": [ + [ + -1.45, + -0.3, + 1.5 + ], + [ + 0.05, + 0.2, + 1.0 + ] + ] + }, + "success_check": true, + "success_check_how": "the video poses the graded episode's own recorded scene states (trajectory.npz frame_state: every asset's root pose and velocity and joint state, recorded during the graded verifier rollout) on the task's own scene build (InteractiveScene.reset_to, no physics step), so it shows the graded run itself. Start state = the frozen instance (instance_ok 1, NVIDIA L40S). The task's success predicate on the posed final state: 0. Verdict (reward.json): success 1, progress 1.0. The verifier's own camera replay of the actions ended on the graded final state: camera_replay_match 1.", + "image": "efe-private:embodiedswe-0.3.0 (verifier tooling: efe:embodiedswe-files-0.3.1)", + "core": true, + "encode": "libx264 preset slow CRF 16, yuv420p, bt709, +faststart", + "poster": "the final frame", + "created": "2026-10-30" +} \ No newline at end of file diff --git a/masters/embodiedswe/puzzle_syringe_bimanual_franka_osc/poster.png b/masters/embodiedswe/puzzle_syringe_bimanual_franka_osc/poster.png new file mode 100644 index 0000000000000000000000000000000000000000..f55165d4a05152ba5d2b62a9e148873e93f9ae3f --- /dev/null +++ b/masters/embodiedswe/puzzle_syringe_bimanual_franka_osc/poster.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:444f6f14535e52dfa0c2ac8bd282c019c9b022b383bf0280843aa5ae2d0e2a16 +size 2725233 diff --git a/masters/embodiedswe/puzzle_syringe_bimanual_franka_osc/video.mp4 b/masters/embodiedswe/puzzle_syringe_bimanual_franka_osc/video.mp4 new file mode 100644 index 0000000000000000000000000000000000000000..86a986f6ac88eb929595843b13bac94ff8e83530 --- /dev/null +++ b/masters/embodiedswe/puzzle_syringe_bimanual_franka_osc/video.mp4 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cb4c5aed73d07af7b4a11c5d3e008477a33b3fee078e3f9b09f8870c5eb018bd +size 11295467 diff --git a/masters/embodiedswe/teaser.mp4 b/masters/embodiedswe/teaser.mp4 new file mode 100644 index 0000000000000000000000000000000000000000..6f08c1842bc6b0b65af813a16b1586cb4deb1de7 --- /dev/null +++ b/masters/embodiedswe/teaser.mp4 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6d83c42fb7c70307dc280fc80b928e6cb8e4801de73021fcb539bcde954b3a10 +size 11038759 diff --git a/masters/embodiedswe/teaser_poster.png b/masters/embodiedswe/teaser_poster.png new file mode 100644 index 0000000000000000000000000000000000000000..38c9f0b76e60107ca857e5b58c70e597c17f57ea --- /dev/null +++ b/masters/embodiedswe/teaser_poster.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7029177245f6272ace60bdf39ea0724845649799b7e2dae4c7eabf5e582eda46 +size 3340656