{ "automatic_hard_gates_passed": false, "candidate": { "path": "/Users/jiangyuhua/Documents/ChatGPT/PhiAgent-0/outputs/figure-two-robot-blanket-fold-realistic/20260825T145924Z-minimax-h3-v2/remote-results/h200-2/seed-2026082503/candidates/figure-two-robot-blanket-fold-photorealistic-v2/seed-2026082503/candidate.mp4", "sha256": "1701268237cc2f993ab23dca60f26cd18c889b0425945b5ff14257faae96199d" }, "claim_boundary": "MiniMax-H3 outputs are generated camera-pixel proposals depicting full-body humanoid cooperation. They are not recorded real-hardware execution and cannot be promoted to physical PASS without the declared physical evidence.", "decision": "REJECTED_AUTOMATIC_HARD_GATE_FAILURE", "diagnostic_logs": { "first_frame_ssim": "Input #0, mov,mp4,m4a,3gp,3g2,mj2, from '/Users/jiangyuhua/Documents/ChatGPT/PhiAgent-0/outputs/figure-two-robot-blanket-fold-realistic/20260825T145924Z-minimax-h3-v2/remote-results/h200-2/seed-2026082503/candidates/figure-two-robot-blanket-fold-photorealistic-v2/seed-2026082503/candidate.mp4':\n Metadata:\n major_brand : isom\n minor_version : 512\n compatible_brands: isomiso2avc1mp41\n encoder : Lavf58.29.100\n Duration: 00:00:08.00, start: 0.000000, bitrate: 807 kb/s\n Stream #0:0[0x1](und): Video: h264 (High) (avc1 / 0x31637661), yuv420p(progressive), 1024x768, 669 kb/s, 24 fps, 24 tbr, 12288 tbn (default)\n Metadata:\n handler_name : VideoHandler\n Stream #0:1[0x2](und): Audio: aac (LC) (mp4a / 0x6134706D), 32000 Hz, stereo, fltp, 129 kb/s (default)\n Metadata:\n handler_name : SoundHandler\nInput #1, png_pipe, from '/Users/jiangyuhua/Documents/ChatGPT/PhiAgent-0/outputs/figure-two-robot-blanket-fold-realistic/20260825T145924Z-minimax-h3-v2/inputs/initial-frame.png':\n Duration: N/A, bitrate: N/A\n Stream #1:0: Video: png, rgb24(pc, gbr/unknown/unknown), 1024x768, 24 fps, 24 tbr, 24 tbn\nStream mapping:\n Stream #0:0 (h264) -> trim:default\n Stream #1:0 (png) -> setpts:default\n ssim:default -> Stream #0:0 (wrapped_avframe)\n[Parsed_ssim_5 @ 0x93b0267c0] not matching timebases found between first input: 1/12288 and second input 1/24, results may be incorrect!\nOutput #0, null, to 'pipe:':\n Metadata:\n major_brand : isom\n minor_version : 512\n compatible_brands: isomiso2avc1mp41\n encoder : Lavf62.12.102\n Stream #0:0: Video: wrapped_avframe, yuv420p(progressive), 1024x768, q=2-31, 200 kb/s, 24 fps, 24 tbn\n Metadata:\n encoder : Lavc62.28.102 wrapped_avframe\n[Parsed_ssim_5 @ 0x93b0267c0] SSIM Y:0.911176 (10.514685) U:0.987483 (19.025021) V:0.986684 (18.756267) All:0.936478 (11.970783)\n[out#0/null @ 0x93b024780] video:0KiB audio:0KiB subtitle:0KiB other streams:0KiB global headers:0KiB muxing overhead: unknown\nframe= 1 fps=0.0 q=-0.0 Lsize=N/A time=00:00:00.04 bitrate=N/A speed=1.14x elapsed=0:00:00.03 \n", "scene_cut": "Input #0, mov,mp4,m4a,3gp,3g2,mj2, from '/Users/jiangyuhua/Documents/ChatGPT/PhiAgent-0/outputs/figure-two-robot-blanket-fold-realistic/20260825T145924Z-minimax-h3-v2/remote-results/h200-2/seed-2026082503/candidates/figure-two-robot-blanket-fold-photorealistic-v2/seed-2026082503/candidate.mp4':\n Metadata:\n major_brand : isom\n minor_version : 512\n compatible_brands: isomiso2avc1mp41\n encoder : Lavf58.29.100\n Duration: 00:00:08.00, start: 0.000000, bitrate: 807 kb/s\n Stream #0:0[0x1](und): Video: h264 (High) (avc1 / 0x31637661), yuv420p(progressive), 1024x768, 669 kb/s, 24 fps, 24 tbr, 12288 tbn (default)\n Metadata:\n handler_name : VideoHandler\n Stream #0:1[0x2](und): Audio: aac (LC) (mp4a / 0x6134706D), 32000 Hz, stereo, fltp, 129 kb/s (default)\n Metadata:\n handler_name : SoundHandler\nStream mapping:\n Stream #0:0 -> #0:0 (h264 (native) -> wrapped_avframe (native))\n[Parsed_showinfo_1 @ 0x9fac0a4c0] config in time_base: 1/12288, frame_rate: 24/1\n[Parsed_showinfo_1 @ 0x9fac0a4c0] config out time_base: 0/0, frame_rate: 0/0\n[vf#0:0 @ 0x9facc4200] No filtered frames for output stream, trying to initialize anyway.\nOutput #0, null, to 'pipe:':\n Metadata:\n major_brand : isom\n minor_version : 512\n compatible_brands: isomiso2avc1mp41\n encoder : Lavf62.12.102\n Stream #0:0(und): Video: wrapped_avframe, yuv420p(progressive), 1024x768, q=2-31, 200 kb/s, 24 fps, 12288 tbn (default)\n Metadata:\n encoder : Lavc62.28.102 wrapped_avframe\n handler_name : VideoHandler\n[out#0/null @ 0x9fac08840] video:0KiB audio:0KiB subtitle:0KiB other streams:0KiB global headers:0KiB muxing overhead: unknown\n[out#0/null @ 0x9fac08840] Output file is empty, nothing was encoded(check -ss / -t / -frames parameters if used)\nframe= 0 fps=0.0 q=0.0 Lsize=N/A time=N/A bitrate=N/A speed=N/A elapsed=0:00:00.07 \n", "terminal_ssim": "Input #0, mov,mp4,m4a,3gp,3g2,mj2, from '/Users/jiangyuhua/Documents/ChatGPT/PhiAgent-0/outputs/figure-two-robot-blanket-fold-realistic/20260825T145924Z-minimax-h3-v2/remote-results/h200-2/seed-2026082503/candidates/figure-two-robot-blanket-fold-photorealistic-v2/seed-2026082503/candidate.mp4':\n Metadata:\n major_brand : isom\n minor_version : 512\n compatible_brands: isomiso2avc1mp41\n encoder : Lavf58.29.100\n Duration: 00:00:08.00, start: 0.000000, bitrate: 807 kb/s\n Stream #0:0[0x1](und): Video: h264 (High) (avc1 / 0x31637661), yuv420p(progressive), 1024x768, 669 kb/s, 24 fps, 24 tbr, 12288 tbn (default)\n Metadata:\n handler_name : VideoHandler\n Stream #0:1[0x2](und): Audio: aac (LC) (mp4a / 0x6134706D), 32000 Hz, stereo, fltp, 129 kb/s (default)\n Metadata:\n handler_name : SoundHandler\nInput #1, png_pipe, from '/Users/jiangyuhua/Documents/ChatGPT/PhiAgent-0/outputs/figure-two-robot-blanket-fold-realistic/20260825T145924Z-minimax-h3-v2/evaluations/seed-2026082503/final-frame.png':\n Duration: N/A, bitrate: N/A\n Stream #1:0: Video: png, rgb24(pc, gbr/unknown/unknown), 1024x768, 24 fps, 24 tbr, 24 tbn\nStream mapping:\n Stream #0:0 (h264) -> trim:default\n Stream #1:0 (png) -> setpts:default\n ssim:default -> Stream #0:0 (wrapped_avframe)\n[Parsed_ssim_5 @ 0x9bb026ac0] not matching timebases found between first input: 1/12288 and second input 1/24, results may be incorrect!\nOutput #0, null, to 'pipe:':\n Metadata:\n major_brand : isom\n minor_version : 512\n compatible_brands: isomiso2avc1mp41\n encoder : Lavf62.12.102\n Stream #0:0: Video: wrapped_avframe, yuv420p(progressive), 1024x768, q=2-31, 200 kb/s, 24 fps, 24 tbn\n Metadata:\n encoder : Lavc62.28.102 wrapped_avframe\n[Parsed_ssim_5 @ 0x9bb026ac0] SSIM Y:0.997612 (26.219404) U:0.999815 (37.338656) V:0.999870 (38.858865) All:0.998355 (27.839588)\n[out#0/null @ 0x9bb024780] video:5KiB audio:0KiB subtitle:0KiB other streams:0KiB global headers:0KiB muxing overhead: unknown\nframe= 12 fps=0.0 q=-0.0 Lsize=N/A time=00:00:00.50 bitrate=N/A speed=4.62x elapsed=0:00:00.10 \n" }, "evaluated_at": "2026-08-25T15:28:30.003986+00:00", "failed_visual_gate_ids": [ "exact_first_frame_ssim_at_least_0_985" ], "honest_status": "BLOCKED", "metrics": { "background_patch_ssim": { "bed_and_window": 0.942003, "table_front": 0.96382, "upper_center_room": 0.935253 }, "detected_cut_times_seconds": [], "first_frame_ssim": 0.936478, "minimum_background_patch_ssim": 0.935253, "terminal_window_ssim_against_final": 0.998355 }, "physical_promotion": false, "physical_promotion_gates": [ { "evidence": { "reason": "generated RGB cannot establish this physical-execution gate" }, "gate_id": "calibrated_metric_camera", "state": "UNAVAILABLE" }, { "evidence": { "reason": "generated RGB cannot establish this physical-execution gate" }, "gate_id": "exact_robot_joint_trajectories", "state": "UNAVAILABLE" }, { "evidence": { "reason": "generated RGB cannot establish this physical-execution gate" }, "gate_id": "persistent_metric_quilt_geometry", "state": "UNAVAILABLE" }, { "evidence": { "reason": "generated RGB cannot establish this physical-execution gate" }, "gate_id": "force_or_tactile_contact_evidence", "state": "UNAVAILABLE" }, { "evidence": { "reason": "generated RGB cannot establish this physical-execution gate" }, "gate_id": "collision_checked_executable_motion", "state": "UNAVAILABLE" }, { "evidence": { "reason": "generated RGB cannot establish this physical-execution gate" }, "gate_id": "recorded_real_robot_execution", "state": "UNAVAILABLE" } ], "review_packet": { "contact_sheet": "/Users/jiangyuhua/Documents/ChatGPT/PhiAgent-0/outputs/figure-two-robot-blanket-fold-realistic/20260825T145924Z-minimax-h3-v2/evaluations/seed-2026082503/native-review-contact-sheet.jpg", "contact_sheet_sha256": "468d77bcfa4e28e19704b667bb4d7a30e6a5b4f64cdd0ade8426ea9e77cd087f", "frame_indices": [ 0, 12, 34, 48, 82, 96, 115, 140, 158, 175, 191 ], "template": "/Users/jiangyuhua/Documents/ChatGPT/PhiAgent-0/outputs/figure-two-robot-blanket-fold-realistic/20260825T145924Z-minimax-h3-v2/evaluations/seed-2026082503/native-review-template.json" }, "schema_version": "1.0.0", "unavailable_visual_gate_ids": [ "exactly_two_persistent_robot_identities", "four_anatomically_consistent_robot_arms_and_hands", "exactly_one_persistent_quilt", "four_quilt_corners_and_bound_edges_persist_or_reacquire", "quilt_color_pattern_thickness_and_topology_persist", "visible_contact_precedes_every_quilt_motion", "bilateral_fold_motion_onset_within_0_25_seconds", "first_fold_settles_before_regrasp", "gatefold_begins_only_after_regrasp", "lifted_cloth_motion_is_gravity_and_support_consistent", "no_robot_robot_or_robot_table_interpenetration", "terminal_compact_rectangle", "no_text_guides_markers_or_watermarks", "candidate_sha_bound_native_resolution_user_review" ], "visual_acceptance": false, "visual_gates": [ { "evidence": {}, "gate_id": "artifact_hash_binding", "state": "PASS" }, { "evidence": { "runtime_probe": { "capabilities": [ [ 9, 0 ], [ 9, 0 ] ], "cuda": "12.8", "cuda_available": true, "device_count": 2, "device_names": [ "NVIDIA H200", "NVIDIA H200" ], "mapped_cuda13_libraries": [], "sglang": "0.5.17", "sglang_kernel": "0.4.5", "sglang_kernel_common_ops": { "path": "/dev/shm/phiagent-h3-h200-cu128-v2/lib/python3.12/site-packages/sgl_kernel/sm90/common_ops.abi3.so", "sha256": "041ad7991302d3faf726be66411223c6bb8de0210abd85bfa7d7d2eb14bea075" }, "torch": "2.11.0+cu128", "triton": "3.6.0" }, "selected": [ { "compute_capability": "9.0", "compute_processes": [], "driver_version": "570.124.06", "free_mib": 143167, "name": "NVIDIA H200", "physical_index": 0, "total_mib": 143771, "used_mib": 1, "uuid": "GPU-fc1d96a1-dd3e-3e77-c18b-4ae819c7b5d1" }, { "compute_capability": "9.0", "compute_processes": [], "driver_version": "570.124.06", "free_mib": 143167, "name": "NVIDIA H200", "physical_index": 1, "total_mib": 143771, "used_mib": 1, "uuid": "GPU-b220235e-a0f6-465e-c5a8-b240327e441b" } ] }, "gate_id": "gpu_and_runtime_preflight", "state": "PASS" }, { "evidence": { "revision": "42ed227ee7df40d41602854ae760620d6eb651fe", "text_encoder_shards": 14, "transformer_shards": 13 }, "gate_id": "pinned_model_revision", "state": "PASS" }, { "evidence": { "probe": { "format": { "duration": "8.000000" }, "stream": { "avg_frame_rate": "24/1", "codec_name": "h264", "duration": "8.000000", "height": 768, "nb_frames": "192", "nb_read_frames": "192", "r_frame_rate": "24/1", "width": 1024 } } }, "gate_id": "media_contract_1024x768_24fps_192frames", "state": "PASS" }, { "evidence": { "ssim": 0.936478, "threshold": 0.985 }, "gate_id": "exact_first_frame_ssim_at_least_0_985", "state": "FAIL" }, { "evidence": { "detected_cut_times_seconds": [], "limitation": "Automatic scene detection covers hard cuts only. Crossfade, teleport, and local material discontinuity remain native-review requirements.", "scene_threshold": 0.3 }, "gate_id": "single_uninterrupted_shot_no_cut_crossfade_or_teleport", "state": "PASS" }, { "evidence": { "minimum_ssim": 0.935253, "patch_ssim": { "bed_and_window": 0.942003, "table_front": 0.96382, "upper_center_room": 0.935253 }, "threshold": 0.9 }, "gate_id": "fixed_camera_background_and_lighting", "state": "PASS" }, { "evidence": { "reason": "candidate-SHA-bound native-resolution full-video review required" }, "gate_id": "exactly_two_persistent_robot_identities", "state": "UNAVAILABLE" }, { "evidence": { "reason": "candidate-SHA-bound native-resolution full-video review required" }, "gate_id": "four_anatomically_consistent_robot_arms_and_hands", "state": "UNAVAILABLE" }, { "evidence": { "reason": "candidate-SHA-bound native-resolution full-video review required" }, "gate_id": "exactly_one_persistent_quilt", "state": "UNAVAILABLE" }, { "evidence": { "reason": "candidate-SHA-bound native-resolution full-video review required" }, "gate_id": "four_quilt_corners_and_bound_edges_persist_or_reacquire", "state": "UNAVAILABLE" }, { "evidence": { "reason": "candidate-SHA-bound native-resolution full-video review required" }, "gate_id": "quilt_color_pattern_thickness_and_topology_persist", "state": "UNAVAILABLE" }, { "evidence": { "reason": "candidate-SHA-bound native-resolution full-video review required" }, "gate_id": "visible_contact_precedes_every_quilt_motion", "state": "UNAVAILABLE" }, { "evidence": { "reason": "candidate-SHA-bound native-resolution full-video review required" }, "gate_id": "bilateral_fold_motion_onset_within_0_25_seconds", "state": "UNAVAILABLE" }, { "evidence": { "reason": "candidate-SHA-bound native-resolution full-video review required" }, "gate_id": "first_fold_settles_before_regrasp", "state": "UNAVAILABLE" }, { "evidence": { "reason": "candidate-SHA-bound native-resolution full-video review required" }, "gate_id": "gatefold_begins_only_after_regrasp", "state": "UNAVAILABLE" }, { "evidence": { "reason": "candidate-SHA-bound native-resolution full-video review required" }, "gate_id": "lifted_cloth_motion_is_gravity_and_support_consistent", "state": "UNAVAILABLE" }, { "evidence": { "reason": "candidate-SHA-bound native-resolution full-video review required" }, "gate_id": "no_robot_robot_or_robot_table_interpenetration", "state": "UNAVAILABLE" }, { "evidence": { "reason": "candidate-SHA-bound native-resolution full-video review required" }, "gate_id": "terminal_compact_rectangle", "state": "UNAVAILABLE" }, { "evidence": { "conservative_proxy_threshold": 0.985, "limitation": "SSIM is a conservative RGB proxy; the named pixel-RMS gate also requires native review.", "terminal_window_ssim_against_final": 0.998355 }, "gate_id": "terminal_motion_below_1_pixel_rms_for_last_12_frames", "state": "PASS" }, { "evidence": { "reason": "candidate-SHA-bound native-resolution full-video review required" }, "gate_id": "no_text_guides_markers_or_watermarks", "state": "UNAVAILABLE" }, { "evidence": { "reason": "candidate-SHA-bound native-resolution full-video review required" }, "gate_id": "candidate_sha_bound_native_resolution_user_review", "state": "UNAVAILABLE" } ] }