{
  "frames": [
    {
      "file": "frame_000.jpg",
      "time": 0.0,
      "log": "Capturing execution frames: sharp_basic.json\n  [ComfyUI] [comfy-env] [VRAM] Before offload 'MoGeWrapper_1' (631 MB) -> cpu: 6567 / 24122 MB\n  [ComfyUI] [ComfyUI-MoGe2] Health check: ping (timeout=5.0s)...\n  [ComfyUI] [ComfyUI-MoGe2] python: /tmp/comfytest-home/.ce/.pixi/envs/moge2-nodes/bin/python\n  [ComfyUI] [ComfyUI-MoGe2] Health check: failed\n  [ComfyUI] [ComfyUI-MoGe2] sys_path sent to worker: ['/workspaces/Sharp-1914/ComfyUI', '/workspaces/Sharp-1914/ComfyUI/custom_nodes/ComfyUI-MoGe2', '/tmp/comfytest-home/.ce/.pixi/envs/moge2-nodes/lib/python3.1/site-packages', '/workspaces/Sharp-1914/ComfyUI/custom_nodes/ComfyUI-MoGe2']\n  [ComfyUI] [ComfyUI-MoGe2] Socket unhealthy, restarting worker...\n  [ComfyUI] [SubprocessWorker] is_pixi=True, python=/tmp/comfytest-home/.ce/.pixi/envs/moge2-nodes/bin/python\n  [ComfyUI] [comfy-env] Invalidated 1 stale model patchers (will be cleaned up during next unload)\n  [ComfyUI] [SubprocessWorker] launching cmd=['/tmp/comfytest-home/.pixi/bin/pixi', 'run', '--as-is', '--manifest-path', '/tmp/comfytest-home/.ce/pixi.toml', '-e', 'moge2-nodes', 'python', '/tmp/comfyui_pvenv_bmmahm2z/persistent_worker.py', 'abstract://comfy_worker_e79ad090de81']...\n  [ComfyUI] [SubprocessWorker] socket_addr=abstract://comfy_worker_e79ad090de81 (abstract, no filesystem path)\n  [ComfyUI] [worker:ComfyUI-MoGe2] ERROR: Failed to import comfy_kitchen, Error: No module named 'comfy_kitchen', fp8 and fp4 support will not be available.\n  [ComfyUI] [worker:ComfyUI-MoGe2] WARNING: comfy_kitchen does not support stochastic FP8 rounding, please update comfy_kitchen.\n  [ComfyUI] [SubprocessWorker] call_id=3 sending model_to_device\n  [ComfyUI] \u001b[1m\u001b[33m[WARNING]\u001b[0m Model 'MoGeWrapper_1' gone from worker (restarted?), treating as offloaded\n  [ComfyUI] [comfy-env] [VRAM] After offload 'MoGeWrapper_1' (631 MB): 4817 / 24122 MB\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m Using RAM pressure cache."
    },
    {
      "file": "frame_001.jpg",
      "time": 2.3,
      "log": "Capturing execution frames: sharp_basic.json\n  [ComfyUI] [comfy-env] [VRAM] Before offload 'MoGeWrapper_1' (631 MB) -> cpu: 6567 / 24122 MB\n  [ComfyUI] [ComfyUI-MoGe2] Health check: ping (timeout=5.0s)...\n  [ComfyUI] [ComfyUI-MoGe2] python: /tmp/comfytest-home/.ce/.pixi/envs/moge2-nodes/bin/python\n  [ComfyUI] [ComfyUI-MoGe2] Health check: failed\n  [ComfyUI] [ComfyUI-MoGe2] sys_path sent to worker: ['/workspaces/Sharp-1914/ComfyUI', '/workspaces/Sharp-1914/ComfyUI/custom_nodes/ComfyUI-MoGe2', '/tmp/comfytest-home/.ce/.pixi/envs/moge2-nodes/lib/python3.1/site-packages', '/workspaces/Sharp-1914/ComfyUI/custom_nodes/ComfyUI-MoGe2']\n  [ComfyUI] [ComfyUI-MoGe2] Socket unhealthy, restarting worker...\n  [ComfyUI] [SubprocessWorker] is_pixi=True, python=/tmp/comfytest-home/.ce/.pixi/envs/moge2-nodes/bin/python\n  [ComfyUI] [comfy-env] Invalidated 1 stale model patchers (will be cleaned up during next unload)\n  [ComfyUI] [SubprocessWorker] launching cmd=['/tmp/comfytest-home/.pixi/bin/pixi', 'run', '--as-is', '--manifest-path', '/tmp/comfytest-home/.ce/pixi.toml', '-e', 'moge2-nodes', 'python', '/tmp/comfyui_pvenv_bmmahm2z/persistent_worker.py', 'abstract://comfy_worker_e79ad090de81']...\n  [ComfyUI] [SubprocessWorker] socket_addr=abstract://comfy_worker_e79ad090de81 (abstract, no filesystem path)\n  [ComfyUI] [worker:ComfyUI-MoGe2] ERROR: Failed to import comfy_kitchen, Error: No module named 'comfy_kitchen', fp8 and fp4 support will not be available.\n  [ComfyUI] [worker:ComfyUI-MoGe2] WARNING: comfy_kitchen does not support stochastic FP8 rounding, please update comfy_kitchen.\n  [ComfyUI] [SubprocessWorker] call_id=3 sending model_to_device\n  [ComfyUI] \u001b[1m\u001b[33m[WARNING]\u001b[0m Model 'MoGeWrapper_1' gone from worker (restarted?), treating as offloaded\n  [ComfyUI] [comfy-env] [VRAM] After offload 'MoGeWrapper_1' (631 MB): 4817 / 24122 MB\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m Using RAM pressure cache.\n  Frame 1 saved (frame_000.jpg, t=0.0s)\n  Validating workflow...\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m === /validate endpoint called ===\n  Queuing workflow for execution...\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m Received JSON with keys: ['prompt']\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m Prompt has 4 nodes\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m   Node 1: LoadImage\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m     Inputs: {\"image\": \"beach.jpg\"}\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m   Node 2: LoadSharpModel\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m     Inputs: {\"precision\": \"auto\"}\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m   Node 5: SharpPredict\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m     Inputs: {\"focal_length_mm\": 30, \"output_prefix\": \"sharp\", \"model\": [\"2\", 0], \"image\": [\"1\", 0]}\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m   Node 6: PreviewGaussianSharp\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m     Inputs: {\"preview_gaussian\": \"\", \"ply_path\": [\"5\", 0], \"extrinsics\": [\"5\", 1], \"intrinsics\": [\"5\", 2]}\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m Generated prompt_id: 1f141f5e-7c30-4e18-9d61-64bb3fb15caa\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m Calling execution.validate_prompt()...\n  [ComfyUI] [ComfyUI-Sharp] python: /tmp/comfytest-home/.ce/.pixi/envs/sharp-nodes/bin/python\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m Validation result: valid=True\n  [ComfyUI] [ComfyUI-Sharp] sys_path sent to worker: ['/workspaces/Sharp-1914/ComfyUI', '/workspaces/Sharp-1914/ComfyUI/custom_nodes/ComfyUI-Sharp', '/tmp/comfytest-home/.ce/.pixi/envs/sharp-nodes/lib/python3.1/site-packages', '/workspaces/Sharp-1914/ComfyUI/custom_nodes/ComfyUI-Sharp']\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m === Validation PASSED ===\n  [ComfyUI] [SubprocessWorker] is_pixi=True, python=/tmp/comfytest-home/.ce/.pixi/envs/sharp-nodes/bin/python\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m got prompt\n  [ComfyUI] [SubprocessWorker] launching cmd=['/tmp/comfytest-home/.pixi/bin/pixi', 'run', '--as-is', '--manifest-path', '/tmp/comfytest-home/.ce/pixi.toml', '-e', 'sharp-nodes', 'python', '/tmp/comfyui_pvenv_qlpb5zhh/persistent_worker.py', 'abstract://comfy_worker_8bd7d9f62285']...\n  [ComfyUI] [comfy-env] Running LoadSharpModel: precision: 'auto'\n  [ComfyUI] [SubprocessWorker] socket_addr=abstract://comfy_worker_8bd7d9f62285 (abstract, no filesystem path)\n  [ComfyUI] [VRAM] Before LoadSharpModel: 4817 / 24122 MB\n  [ComfyUI] [SubprocessWorker] call_method: nodes.load_model.LoadSharpModel.EXECUTE_NORMALIZED\n  [ComfyUI] [SubprocessWorker] acquired lock, ensuring started...\n  [ComfyUI] [ComfyUI-Sharp] Health check: ping (timeout=5.0s)...\n  [ComfyUI] [ComfyUI-Sharp] Health check: failed\n  [ComfyUI] [ComfyUI-Sharp] Socket unhealthy, restarting worker...\n  [ComfyUI] [worker:ComfyUI-Sharp] ERROR: Failed to import comfy_kitchen, Error: No module named 'comfy_kitchen', fp8 and fp4 support will not be available.\n  [ComfyUI] [worker:ComfyUI-Sharp] WARNING: comfy_kitchen does not support stochastic FP8 rounding, please update comfy_kitchen.\n  [ComfyUI] [SubprocessWorker] worker started/confirmed\n  [ComfyUI] [SubprocessWorker] serializing kwargs to shm...\n  [ComfyUI] [SubprocessWorker] created 0 shm blocks\n  [ComfyUI] [SubprocessWorker] sending request via socket...\n  [ComfyUI] [SubprocessWorker] call_id=2 sending call_method\n  [ComfyUI] [worker:ComfyUI-Sharp] [LoadSharpModel] precision='auto' -> dtype=bf16 (device=cuda:0)\n  [ComfyUI] [worker:ComfyUI-Sharp] [LoadSharpModel] checkpoint already present, reusing (no download): /workspaces/Sharp-1914/ComfyUI/models/sharp/sharp_2572gikvuh.pt (2679.6 MB)\n  [ComfyUI] [worker:ComfyUI-Sharp] [LoadSharpModel] checkpoint ready: /workspaces/Sharp-1914/ComfyUI/models/sharp/sharp_2572gikvuh.pt (2679.6 MB)\n  [ComfyUI] [worker:ComfyUI-Sharp] [LoadSharpModel] config: precision=auto -> dtype=bf16\n  [ComfyUI] [SubprocessWorker] got response: ok\n  [ComfyUI] [comfy-env] [VRAM] Before offload 'MoGeWrapper_1' (631 MB) -> cpu: 4817 / 24122 MB\n  [ComfyUI] [ComfyUI-MoGe2] Health check: ping (timeout=5.0s)...\n  [ComfyUI] [ComfyUI-MoGe2] Health check: ok\n  [ComfyUI] [SubprocessWorker] call_id=4 sending model_to_device\n  [ComfyUI] \u001b[1m\u001b[33m[WARNING]\u001b[0m Model 'MoGeWrapper_1' gone from worker (restarted?), treating as offloaded\n  [ComfyUI] [comfy-env] [VRAM] After offload 'MoGeWrapper_1' (631 MB): 4817 / 24122 MB\n  [ComfyUI] [comfy-env] LoadSharpModel done (1.64s): result: NodeOutput\n  [ComfyUI] [VRAM] After LoadSharpModel: 4817 / 24122 MB\n  [ComfyUI] [comfy-env] Running SharpPredict: focal_length_mm: 30.0, output_prefix: 'sharp', model: dict, image: torch.float32 [1,1024,1536,3] cpu\n  [ComfyUI] [VRAM] Before SharpPredict: 4817 / 24122 MB\n  [ComfyUI] [SubprocessWorker] call_method: nodes.predict.SharpPredict.EXECUTE_NORMALIZED\n  [ComfyUI] [SubprocessWorker] acquired lock, ensuring started...\n  [ComfyUI] [ComfyUI-Sharp] Health check: ping (timeout=5.0s)...\n  [ComfyUI] [ComfyUI-Sharp] Health check: ok\n  [ComfyUI] [SubprocessWorker] worker started/confirmed\n  [ComfyUI] [comfy-env] PRE-SERIALIZE 'image' shape: torch.Size([1, 1024, 1536, 3])\n  [ComfyUI] [SubprocessWorker] serializing kwargs to shm...\n  [ComfyUI] [SubprocessWorker] created 0 shm blocks\n  [ComfyUI] [SubprocessWorker] sending request via socket...\n  [ComfyUI] [SubprocessWorker] call_id=3 sending call_method\n  [ComfyUI] [worker:ComfyUI-Sharp] [comfy-env] _from_shm got dict with keys: ['__type__', 'strategy', 'parent_pid', 'fd', 'storage_size']\n  [ComfyUI] [worker:ComfyUI-Sharp] [comfy-env] DESERIALIZE TensorRef: tensor_size=[1, 1024, 1536, 3]\n  [ComfyUI] [worker:ComfyUI-Sharp] [comfy-env] DESERIALIZED tensor shape: torch.Size([1, 1024, 1536, 3])"
    },
    {
      "file": "frame_002.jpg",
      "time": 4.51,
      "log": "Capturing execution frames: sharp_basic.json\n  [ComfyUI] [comfy-env] [VRAM] Before offload 'MoGeWrapper_1' (631 MB) -> cpu: 6567 / 24122 MB\n  [ComfyUI] [ComfyUI-MoGe2] Health check: ping (timeout=5.0s)...\n  [ComfyUI] [ComfyUI-MoGe2] python: /tmp/comfytest-home/.ce/.pixi/envs/moge2-nodes/bin/python\n  [ComfyUI] [ComfyUI-MoGe2] Health check: failed\n  [ComfyUI] [ComfyUI-MoGe2] sys_path sent to worker: ['/workspaces/Sharp-1914/ComfyUI', '/workspaces/Sharp-1914/ComfyUI/custom_nodes/ComfyUI-MoGe2', '/tmp/comfytest-home/.ce/.pixi/envs/moge2-nodes/lib/python3.1/site-packages', '/workspaces/Sharp-1914/ComfyUI/custom_nodes/ComfyUI-MoGe2']\n  [ComfyUI] [ComfyUI-MoGe2] Socket unhealthy, restarting worker...\n  [ComfyUI] [SubprocessWorker] is_pixi=True, python=/tmp/comfytest-home/.ce/.pixi/envs/moge2-nodes/bin/python\n  [ComfyUI] [comfy-env] Invalidated 1 stale model patchers (will be cleaned up during next unload)\n  [ComfyUI] [SubprocessWorker] launching cmd=['/tmp/comfytest-home/.pixi/bin/pixi', 'run', '--as-is', '--manifest-path', '/tmp/comfytest-home/.ce/pixi.toml', '-e', 'moge2-nodes', 'python', '/tmp/comfyui_pvenv_bmmahm2z/persistent_worker.py', 'abstract://comfy_worker_e79ad090de81']...\n  [ComfyUI] [SubprocessWorker] socket_addr=abstract://comfy_worker_e79ad090de81 (abstract, no filesystem path)\n  [ComfyUI] [worker:ComfyUI-MoGe2] ERROR: Failed to import comfy_kitchen, Error: No module named 'comfy_kitchen', fp8 and fp4 support will not be available.\n  [ComfyUI] [worker:ComfyUI-MoGe2] WARNING: comfy_kitchen does not support stochastic FP8 rounding, please update comfy_kitchen.\n  [ComfyUI] [SubprocessWorker] call_id=3 sending model_to_device\n  [ComfyUI] \u001b[1m\u001b[33m[WARNING]\u001b[0m Model 'MoGeWrapper_1' gone from worker (restarted?), treating as offloaded\n  [ComfyUI] [comfy-env] [VRAM] After offload 'MoGeWrapper_1' (631 MB): 4817 / 24122 MB\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m Using RAM pressure cache.\n  Frame 1 saved (frame_000.jpg, t=0.0s)\n  Validating workflow...\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m === /validate endpoint called ===\n  Queuing workflow for execution...\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m Received JSON with keys: ['prompt']\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m Prompt has 4 nodes\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m   Node 1: LoadImage\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m     Inputs: {\"image\": \"beach.jpg\"}\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m   Node 2: LoadSharpModel\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m     Inputs: {\"precision\": \"auto\"}\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m   Node 5: SharpPredict\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m     Inputs: {\"focal_length_mm\": 30, \"output_prefix\": \"sharp\", \"model\": [\"2\", 0], \"image\": [\"1\", 0]}\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m   Node 6: PreviewGaussianSharp\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m     Inputs: {\"preview_gaussian\": \"\", \"ply_path\": [\"5\", 0], \"extrinsics\": [\"5\", 1], \"intrinsics\": [\"5\", 2]}\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m Generated prompt_id: 1f141f5e-7c30-4e18-9d61-64bb3fb15caa\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m Calling execution.validate_prompt()...\n  [ComfyUI] [ComfyUI-Sharp] python: /tmp/comfytest-home/.ce/.pixi/envs/sharp-nodes/bin/python\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m Validation result: valid=True\n  [ComfyUI] [ComfyUI-Sharp] sys_path sent to worker: ['/workspaces/Sharp-1914/ComfyUI', '/workspaces/Sharp-1914/ComfyUI/custom_nodes/ComfyUI-Sharp', '/tmp/comfytest-home/.ce/.pixi/envs/sharp-nodes/lib/python3.1/site-packages', '/workspaces/Sharp-1914/ComfyUI/custom_nodes/ComfyUI-Sharp']\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m === Validation PASSED ===\n  [ComfyUI] [SubprocessWorker] is_pixi=True, python=/tmp/comfytest-home/.ce/.pixi/envs/sharp-nodes/bin/python\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m got prompt\n  [ComfyUI] [SubprocessWorker] launching cmd=['/tmp/comfytest-home/.pixi/bin/pixi', 'run', '--as-is', '--manifest-path', '/tmp/comfytest-home/.ce/pixi.toml', '-e', 'sharp-nodes', 'python', '/tmp/comfyui_pvenv_qlpb5zhh/persistent_worker.py', 'abstract://comfy_worker_8bd7d9f62285']...\n  [ComfyUI] [comfy-env] Running LoadSharpModel: precision: 'auto'\n  [ComfyUI] [SubprocessWorker] socket_addr=abstract://comfy_worker_8bd7d9f62285 (abstract, no filesystem path)\n  [ComfyUI] [VRAM] Before LoadSharpModel: 4817 / 24122 MB\n  [ComfyUI] [SubprocessWorker] call_method: nodes.load_model.LoadSharpModel.EXECUTE_NORMALIZED\n  [ComfyUI] [SubprocessWorker] acquired lock, ensuring started...\n  [ComfyUI] [ComfyUI-Sharp] Health check: ping (timeout=5.0s)...\n  [ComfyUI] [ComfyUI-Sharp] Health check: failed\n  [ComfyUI] [ComfyUI-Sharp] Socket unhealthy, restarting worker...\n  [ComfyUI] [worker:ComfyUI-Sharp] ERROR: Failed to import comfy_kitchen, Error: No module named 'comfy_kitchen', fp8 and fp4 support will not be available.\n  [ComfyUI] [worker:ComfyUI-Sharp] WARNING: comfy_kitchen does not support stochastic FP8 rounding, please update comfy_kitchen.\n  [ComfyUI] [SubprocessWorker] worker started/confirmed\n  [ComfyUI] [SubprocessWorker] serializing kwargs to shm...\n  [ComfyUI] [SubprocessWorker] created 0 shm blocks\n  [ComfyUI] [SubprocessWorker] sending request via socket...\n  [ComfyUI] [SubprocessWorker] call_id=2 sending call_method\n  [ComfyUI] [worker:ComfyUI-Sharp] [LoadSharpModel] precision='auto' -> dtype=bf16 (device=cuda:0)\n  [ComfyUI] [worker:ComfyUI-Sharp] [LoadSharpModel] checkpoint already present, reusing (no download): /workspaces/Sharp-1914/ComfyUI/models/sharp/sharp_2572gikvuh.pt (2679.6 MB)\n  [ComfyUI] [worker:ComfyUI-Sharp] [LoadSharpModel] checkpoint ready: /workspaces/Sharp-1914/ComfyUI/models/sharp/sharp_2572gikvuh.pt (2679.6 MB)\n  [ComfyUI] [worker:ComfyUI-Sharp] [LoadSharpModel] config: precision=auto -> dtype=bf16\n  [ComfyUI] [SubprocessWorker] got response: ok\n  [ComfyUI] [comfy-env] [VRAM] Before offload 'MoGeWrapper_1' (631 MB) -> cpu: 4817 / 24122 MB\n  [ComfyUI] [ComfyUI-MoGe2] Health check: ping (timeout=5.0s)...\n  [ComfyUI] [ComfyUI-MoGe2] Health check: ok\n  [ComfyUI] [SubprocessWorker] call_id=4 sending model_to_device\n  [ComfyUI] \u001b[1m\u001b[33m[WARNING]\u001b[0m Model 'MoGeWrapper_1' gone from worker (restarted?), treating as offloaded\n  [ComfyUI] [comfy-env] [VRAM] After offload 'MoGeWrapper_1' (631 MB): 4817 / 24122 MB\n  [ComfyUI] [comfy-env] LoadSharpModel done (1.64s): result: NodeOutput\n  [ComfyUI] [VRAM] After LoadSharpModel: 4817 / 24122 MB\n  [ComfyUI] [comfy-env] Running SharpPredict: focal_length_mm: 30.0, output_prefix: 'sharp', model: dict, image: torch.float32 [1,1024,1536,3] cpu\n  [ComfyUI] [VRAM] Before SharpPredict: 4817 / 24122 MB\n  [ComfyUI] [SubprocessWorker] call_method: nodes.predict.SharpPredict.EXECUTE_NORMALIZED\n  [ComfyUI] [SubprocessWorker] acquired lock, ensuring started...\n  [ComfyUI] [ComfyUI-Sharp] Health check: ping (timeout=5.0s)...\n  [ComfyUI] [ComfyUI-Sharp] Health check: ok\n  [ComfyUI] [SubprocessWorker] worker started/confirmed\n  [ComfyUI] [comfy-env] PRE-SERIALIZE 'image' shape: torch.Size([1, 1024, 1536, 3])\n  [ComfyUI] [SubprocessWorker] serializing kwargs to shm...\n  [ComfyUI] [SubprocessWorker] created 0 shm blocks\n  [ComfyUI] [SubprocessWorker] sending request via socket...\n  [ComfyUI] [SubprocessWorker] call_id=3 sending call_method\n  [ComfyUI] [worker:ComfyUI-Sharp] [comfy-env] _from_shm got dict with keys: ['__type__', 'strategy', 'parent_pid', 'fd', 'storage_size']\n  [ComfyUI] [worker:ComfyUI-Sharp] [comfy-env] DESERIALIZE TensorRef: tensor_size=[1, 1024, 1536, 3]\n  [ComfyUI] [worker:ComfyUI-Sharp] [comfy-env] DESERIALIZED tensor shape: torch.Size([1, 1024, 1536, 3])\n  Frame 2 saved (frame_001.jpg, t=2.3s)\n  [capture-periodic] freeze=1ms shot=108ms unfreeze=3ms saved=True"
    },
    {
      "file": "frame_003.jpg",
      "time": 15.39,
      "log": "Capturing execution frames: sharp_basic.json\n  [ComfyUI] [comfy-env] [VRAM] Before offload 'MoGeWrapper_1' (631 MB) -> cpu: 6567 / 24122 MB\n  [ComfyUI] [ComfyUI-MoGe2] Health check: ping (timeout=5.0s)...\n  [ComfyUI] [ComfyUI-MoGe2] python: /tmp/comfytest-home/.ce/.pixi/envs/moge2-nodes/bin/python\n  [ComfyUI] [ComfyUI-MoGe2] Health check: failed\n  [ComfyUI] [ComfyUI-MoGe2] sys_path sent to worker: ['/workspaces/Sharp-1914/ComfyUI', '/workspaces/Sharp-1914/ComfyUI/custom_nodes/ComfyUI-MoGe2', '/tmp/comfytest-home/.ce/.pixi/envs/moge2-nodes/lib/python3.1/site-packages', '/workspaces/Sharp-1914/ComfyUI/custom_nodes/ComfyUI-MoGe2']\n  [ComfyUI] [ComfyUI-MoGe2] Socket unhealthy, restarting worker...\n  [ComfyUI] [SubprocessWorker] is_pixi=True, python=/tmp/comfytest-home/.ce/.pixi/envs/moge2-nodes/bin/python\n  [ComfyUI] [comfy-env] Invalidated 1 stale model patchers (will be cleaned up during next unload)\n  [ComfyUI] [SubprocessWorker] launching cmd=['/tmp/comfytest-home/.pixi/bin/pixi', 'run', '--as-is', '--manifest-path', '/tmp/comfytest-home/.ce/pixi.toml', '-e', 'moge2-nodes', 'python', '/tmp/comfyui_pvenv_bmmahm2z/persistent_worker.py', 'abstract://comfy_worker_e79ad090de81']...\n  [ComfyUI] [SubprocessWorker] socket_addr=abstract://comfy_worker_e79ad090de81 (abstract, no filesystem path)\n  [ComfyUI] [worker:ComfyUI-MoGe2] ERROR: Failed to import comfy_kitchen, Error: No module named 'comfy_kitchen', fp8 and fp4 support will not be available.\n  [ComfyUI] [worker:ComfyUI-MoGe2] WARNING: comfy_kitchen does not support stochastic FP8 rounding, please update comfy_kitchen.\n  [ComfyUI] [SubprocessWorker] call_id=3 sending model_to_device\n  [ComfyUI] \u001b[1m\u001b[33m[WARNING]\u001b[0m Model 'MoGeWrapper_1' gone from worker (restarted?), treating as offloaded\n  [ComfyUI] [comfy-env] [VRAM] After offload 'MoGeWrapper_1' (631 MB): 4817 / 24122 MB\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m Using RAM pressure cache.\n  Frame 1 saved (frame_000.jpg, t=0.0s)\n  Validating workflow...\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m === /validate endpoint called ===\n  Queuing workflow for execution...\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m Received JSON with keys: ['prompt']\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m Prompt has 4 nodes\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m   Node 1: LoadImage\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m     Inputs: {\"image\": \"beach.jpg\"}\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m   Node 2: LoadSharpModel\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m     Inputs: {\"precision\": \"auto\"}\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m   Node 5: SharpPredict\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m     Inputs: {\"focal_length_mm\": 30, \"output_prefix\": \"sharp\", \"model\": [\"2\", 0], \"image\": [\"1\", 0]}\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m   Node 6: PreviewGaussianSharp\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m     Inputs: {\"preview_gaussian\": \"\", \"ply_path\": [\"5\", 0], \"extrinsics\": [\"5\", 1], \"intrinsics\": [\"5\", 2]}\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m Generated prompt_id: 1f141f5e-7c30-4e18-9d61-64bb3fb15caa\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m Calling execution.validate_prompt()...\n  [ComfyUI] [ComfyUI-Sharp] python: /tmp/comfytest-home/.ce/.pixi/envs/sharp-nodes/bin/python\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m Validation result: valid=True\n  [ComfyUI] [ComfyUI-Sharp] sys_path sent to worker: ['/workspaces/Sharp-1914/ComfyUI', '/workspaces/Sharp-1914/ComfyUI/custom_nodes/ComfyUI-Sharp', '/tmp/comfytest-home/.ce/.pixi/envs/sharp-nodes/lib/python3.1/site-packages', '/workspaces/Sharp-1914/ComfyUI/custom_nodes/ComfyUI-Sharp']\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m === Validation PASSED ===\n  [ComfyUI] [SubprocessWorker] is_pixi=True, python=/tmp/comfytest-home/.ce/.pixi/envs/sharp-nodes/bin/python\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m got prompt\n  [ComfyUI] [SubprocessWorker] launching cmd=['/tmp/comfytest-home/.pixi/bin/pixi', 'run', '--as-is', '--manifest-path', '/tmp/comfytest-home/.ce/pixi.toml', '-e', 'sharp-nodes', 'python', '/tmp/comfyui_pvenv_qlpb5zhh/persistent_worker.py', 'abstract://comfy_worker_8bd7d9f62285']...\n  [ComfyUI] [comfy-env] Running LoadSharpModel: precision: 'auto'\n  [ComfyUI] [SubprocessWorker] socket_addr=abstract://comfy_worker_8bd7d9f62285 (abstract, no filesystem path)\n  [ComfyUI] [VRAM] Before LoadSharpModel: 4817 / 24122 MB\n  [ComfyUI] [SubprocessWorker] call_method: nodes.load_model.LoadSharpModel.EXECUTE_NORMALIZED\n  [ComfyUI] [SubprocessWorker] acquired lock, ensuring started...\n  [ComfyUI] [ComfyUI-Sharp] Health check: ping (timeout=5.0s)...\n  [ComfyUI] [ComfyUI-Sharp] Health check: failed\n  [ComfyUI] [ComfyUI-Sharp] Socket unhealthy, restarting worker...\n  [ComfyUI] [worker:ComfyUI-Sharp] ERROR: Failed to import comfy_kitchen, Error: No module named 'comfy_kitchen', fp8 and fp4 support will not be available.\n  [ComfyUI] [worker:ComfyUI-Sharp] WARNING: comfy_kitchen does not support stochastic FP8 rounding, please update comfy_kitchen.\n  [ComfyUI] [SubprocessWorker] worker started/confirmed\n  [ComfyUI] [SubprocessWorker] serializing kwargs to shm...\n  [ComfyUI] [SubprocessWorker] created 0 shm blocks\n  [ComfyUI] [SubprocessWorker] sending request via socket...\n  [ComfyUI] [SubprocessWorker] call_id=2 sending call_method\n  [ComfyUI] [worker:ComfyUI-Sharp] [LoadSharpModel] precision='auto' -> dtype=bf16 (device=cuda:0)\n  [ComfyUI] [worker:ComfyUI-Sharp] [LoadSharpModel] checkpoint already present, reusing (no download): /workspaces/Sharp-1914/ComfyUI/models/sharp/sharp_2572gikvuh.pt (2679.6 MB)\n  [ComfyUI] [worker:ComfyUI-Sharp] [LoadSharpModel] checkpoint ready: /workspaces/Sharp-1914/ComfyUI/models/sharp/sharp_2572gikvuh.pt (2679.6 MB)\n  [ComfyUI] [worker:ComfyUI-Sharp] [LoadSharpModel] config: precision=auto -> dtype=bf16\n  [ComfyUI] [SubprocessWorker] got response: ok\n  [ComfyUI] [comfy-env] [VRAM] Before offload 'MoGeWrapper_1' (631 MB) -> cpu: 4817 / 24122 MB\n  [ComfyUI] [ComfyUI-MoGe2] Health check: ping (timeout=5.0s)...\n  [ComfyUI] [ComfyUI-MoGe2] Health check: ok\n  [ComfyUI] [SubprocessWorker] call_id=4 sending model_to_device\n  [ComfyUI] \u001b[1m\u001b[33m[WARNING]\u001b[0m Model 'MoGeWrapper_1' gone from worker (restarted?), treating as offloaded\n  [ComfyUI] [comfy-env] [VRAM] After offload 'MoGeWrapper_1' (631 MB): 4817 / 24122 MB\n  [ComfyUI] [comfy-env] LoadSharpModel done (1.64s): result: NodeOutput\n  [ComfyUI] [VRAM] After LoadSharpModel: 4817 / 24122 MB\n  [ComfyUI] [comfy-env] Running SharpPredict: focal_length_mm: 30.0, output_prefix: 'sharp', model: dict, image: torch.float32 [1,1024,1536,3] cpu\n  [ComfyUI] [VRAM] Before SharpPredict: 4817 / 24122 MB\n  [ComfyUI] [SubprocessWorker] call_method: nodes.predict.SharpPredict.EXECUTE_NORMALIZED\n  [ComfyUI] [SubprocessWorker] acquired lock, ensuring started...\n  [ComfyUI] [ComfyUI-Sharp] Health check: ping (timeout=5.0s)...\n  [ComfyUI] [ComfyUI-Sharp] Health check: ok\n  [ComfyUI] [SubprocessWorker] worker started/confirmed\n  [ComfyUI] [comfy-env] PRE-SERIALIZE 'image' shape: torch.Size([1, 1024, 1536, 3])\n  [ComfyUI] [SubprocessWorker] serializing kwargs to shm...\n  [ComfyUI] [SubprocessWorker] created 0 shm blocks\n  [ComfyUI] [SubprocessWorker] sending request via socket...\n  [ComfyUI] [SubprocessWorker] call_id=3 sending call_method\n  [ComfyUI] [worker:ComfyUI-Sharp] [comfy-env] _from_shm got dict with keys: ['__type__', 'strategy', 'parent_pid', 'fd', 'storage_size']\n  [ComfyUI] [worker:ComfyUI-Sharp] [comfy-env] DESERIALIZE TensorRef: tensor_size=[1, 1024, 1536, 3]\n  [ComfyUI] [worker:ComfyUI-Sharp] [comfy-env] DESERIALIZED tensor shape: torch.Size([1, 1024, 1536, 3])\n  Frame 2 saved (frame_001.jpg, t=2.3s)\n  [capture-periodic] freeze=1ms shot=108ms unfreeze=3ms saved=True\n  Frame 3 saved (frame_002.jpg, t=4.5s)\n  [capture-periodic] freeze=1ms shot=108ms unfreeze=3ms saved=True\n  [capture-loop] iter=50 t=5.6s state={'complete': False, 'error': None, 'executedCount': 0, 'wsState': 1} eval_ms=1 ram=10.8/31.3GB (34%) swap=1.8/32.0GB (6%) cpu=15% browser_rss=1.19GB\n  [ComfyUI] [worker:ComfyUI-Sharp] [LoadSharpModel] loading checkpoint from /workspaces/Sharp-1914/ComfyUI/models/sharp/sharp_2572gikvuh.pt (dtype=bf16)\n  [ComfyUI] [worker:ComfyUI-Sharp] [LoadSharpModel] initializing model on meta device...\n  [ComfyUI] [worker:ComfyUI-Sharp] [LoadSharpModel] model ready (torch.bfloat16)\n  [ComfyUI] [SubprocessWorker] callback 'request_vram_budget' from call_id=3\n  [ComfyUI] [comfy-env] VRAM request: 2.81GB, free before eviction: 20.24GB\n  [ComfyUI] [comfy-env] VRAM after eviction: 20.24GB\n  [ComfyUI] [worker:ComfyUI-Sharp] [VRAM] +715MB (now 715MB) reserved=736MB peak=715MB\n  [ComfyUI] [worker:ComfyUI-Sharp] [VRAM] +2563MB (now 3278MB) reserved=3456MB peak=3350MB\n  [ComfyUI] [worker:ComfyUI-Sharp] [VRAM] +415MB (now 3693MB) reserved=4032MB peak=3693MB\n  [capture-loop] iter=100 t=11.0s state={'complete': False, 'error': None, 'executedCount': 0, 'wsState': 1} eval_ms=2 ram=11.1/31.3GB (36%) swap=1.8/32.0GB (6%) cpu=13% browser_rss=1.09GB\n  [ComfyUI] [worker:ComfyUI-Sharp] [VRAM] +470MB (now 4164MB) reserved=7840MB peak=7053MB\n  [ComfyUI] [worker:ComfyUI-Sharp] [VRAM] +234MB (now 4398MB) reserved=7360MB peak=7053MB\n  [ComfyUI] [worker:ComfyUI-Sharp] [VRAM] -1544MB (now 2854MB) reserved=7360MB peak=7053MB\n  [ComfyUI] [SubprocessWorker] callback 'report_progress' from call_id=3\n  [ComfyUI] [SubprocessWorker] got response: ok\n  [ComfyUI] [comfy-env] SubprocessWorker: /tmp/comfytest-home/.ce/.pixi/envs/gaussianpack-nodes/bin/python\n  [ComfyUI] [comfy-env] env_vars: {'COMFYUI_BASE': '/workspaces/Sharp-1914/ComfyUI'}\n  [ComfyUI] [comfy-env] Created 1 model patchers: ['RGBGaussianPredictor_1']\n  [ComfyUI] [ComfyUI-GaussianPack] python: /tmp/comfytest-home/.ce/.pixi/envs/gaussianpack-nodes/bin/python\n  [ComfyUI] [comfy-env] SharpPredict done (12.18s): result: NodeOutput\n  [ComfyUI] [VRAM] After SharpPredict: 12291 / 24122 MB\n  [ComfyUI] [ComfyUI-GaussianPack] sys_path sent to worker: ['/workspaces/Sharp-1914/ComfyUI', '/workspaces/Sharp-1914/ComfyUI/custom_nodes/ComfyUI-GaussianPack', '/tmp/comfytest-home/.ce/.pixi/envs/gaussianpack-nodes/lib/python3.1/site-packages', '/workspaces/Sharp-1914/ComfyUI/custom_nodes/ComfyUI-GaussianPack']\n  [ComfyUI] [VRAM] Loaded: SubprocessModel (2679 MB)\n  [ComfyUI] [SubprocessWorker] is_pixi=True, python=/tmp/comfytest-home/.ce/.pixi/envs/gaussianpack-nodes/bin/python\n  [ComfyUI] [comfy-env] Running PreviewGaussianSharp: ply_path: '/workspaces/Sharp-1914/ComfyUI/output/sharp_178086030688..., extrinsics: list[4], intrinsics: list[3]\n  [ComfyUI] [SubprocessWorker] launching cmd=['/tmp/comfytest-home/.pixi/bin/pixi', 'run', '--as-is', '--manifest-path', '/tmp/comfytest-home/.ce/pixi.toml', '-e', 'gaussianpack-nodes', 'python', '/tmp/comfyui_pvenv_ii0358nh/persistent_worker.py', 'abstract://comfy_worker_26954ef0c0d3']...\n  [ComfyUI] [VRAM] Before PreviewGaussianSharp: 12291 / 24122 MB\n  [ComfyUI] [SubprocessWorker] socket_addr=abstract://comfy_worker_26954ef0c0d3 (abstract, no filesystem path)\n  [ComfyUI] [VRAM] Loaded: SubprocessModel (2679 MB)\n  [ComfyUI] [SubprocessWorker] call_method: nodes.preview_gaussian_sharp.PreviewGaussianSharp.EXECUTE_NORMALIZED\n  [ComfyUI] [SubprocessWorker] acquired lock, ensuring started...\n  [ComfyUI] [SubprocessWorker] worker started/confirmed\n  [ComfyUI] [SubprocessWorker] serializing kwargs to shm...\n  [ComfyUI] [SubprocessWorker] created 0 shm blocks\n  [ComfyUI] [SubprocessWorker] sending request via socket...\n  [ComfyUI] [SubprocessWorker] call_id=1 sending call_method"
    },
    {
      "file": "frame_004.jpg",
      "time": 15.88,
      "log": "Capturing execution frames: sharp_basic.json\n  [ComfyUI] [comfy-env] [VRAM] Before offload 'MoGeWrapper_1' (631 MB) -> cpu: 6567 / 24122 MB\n  [ComfyUI] [ComfyUI-MoGe2] Health check: ping (timeout=5.0s)...\n  [ComfyUI] [ComfyUI-MoGe2] python: /tmp/comfytest-home/.ce/.pixi/envs/moge2-nodes/bin/python\n  [ComfyUI] [ComfyUI-MoGe2] Health check: failed\n  [ComfyUI] [ComfyUI-MoGe2] sys_path sent to worker: ['/workspaces/Sharp-1914/ComfyUI', '/workspaces/Sharp-1914/ComfyUI/custom_nodes/ComfyUI-MoGe2', '/tmp/comfytest-home/.ce/.pixi/envs/moge2-nodes/lib/python3.1/site-packages', '/workspaces/Sharp-1914/ComfyUI/custom_nodes/ComfyUI-MoGe2']\n  [ComfyUI] [ComfyUI-MoGe2] Socket unhealthy, restarting worker...\n  [ComfyUI] [SubprocessWorker] is_pixi=True, python=/tmp/comfytest-home/.ce/.pixi/envs/moge2-nodes/bin/python\n  [ComfyUI] [comfy-env] Invalidated 1 stale model patchers (will be cleaned up during next unload)\n  [ComfyUI] [SubprocessWorker] launching cmd=['/tmp/comfytest-home/.pixi/bin/pixi', 'run', '--as-is', '--manifest-path', '/tmp/comfytest-home/.ce/pixi.toml', '-e', 'moge2-nodes', 'python', '/tmp/comfyui_pvenv_bmmahm2z/persistent_worker.py', 'abstract://comfy_worker_e79ad090de81']...\n  [ComfyUI] [SubprocessWorker] socket_addr=abstract://comfy_worker_e79ad090de81 (abstract, no filesystem path)\n  [ComfyUI] [worker:ComfyUI-MoGe2] ERROR: Failed to import comfy_kitchen, Error: No module named 'comfy_kitchen', fp8 and fp4 support will not be available.\n  [ComfyUI] [worker:ComfyUI-MoGe2] WARNING: comfy_kitchen does not support stochastic FP8 rounding, please update comfy_kitchen.\n  [ComfyUI] [SubprocessWorker] call_id=3 sending model_to_device\n  [ComfyUI] \u001b[1m\u001b[33m[WARNING]\u001b[0m Model 'MoGeWrapper_1' gone from worker (restarted?), treating as offloaded\n  [ComfyUI] [comfy-env] [VRAM] After offload 'MoGeWrapper_1' (631 MB): 4817 / 24122 MB\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m Using RAM pressure cache.\n  Frame 1 saved (frame_000.jpg, t=0.0s)\n  Validating workflow...\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m === /validate endpoint called ===\n  Queuing workflow for execution...\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m Received JSON with keys: ['prompt']\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m Prompt has 4 nodes\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m   Node 1: LoadImage\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m     Inputs: {\"image\": \"beach.jpg\"}\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m   Node 2: LoadSharpModel\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m     Inputs: {\"precision\": \"auto\"}\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m   Node 5: SharpPredict\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m     Inputs: {\"focal_length_mm\": 30, \"output_prefix\": \"sharp\", \"model\": [\"2\", 0], \"image\": [\"1\", 0]}\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m   Node 6: PreviewGaussianSharp\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m     Inputs: {\"preview_gaussian\": \"\", \"ply_path\": [\"5\", 0], \"extrinsics\": [\"5\", 1], \"intrinsics\": [\"5\", 2]}\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m Generated prompt_id: 1f141f5e-7c30-4e18-9d61-64bb3fb15caa\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m Calling execution.validate_prompt()...\n  [ComfyUI] [ComfyUI-Sharp] python: /tmp/comfytest-home/.ce/.pixi/envs/sharp-nodes/bin/python\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m Validation result: valid=True\n  [ComfyUI] [ComfyUI-Sharp] sys_path sent to worker: ['/workspaces/Sharp-1914/ComfyUI', '/workspaces/Sharp-1914/ComfyUI/custom_nodes/ComfyUI-Sharp', '/tmp/comfytest-home/.ce/.pixi/envs/sharp-nodes/lib/python3.1/site-packages', '/workspaces/Sharp-1914/ComfyUI/custom_nodes/ComfyUI-Sharp']\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m === Validation PASSED ===\n  [ComfyUI] [SubprocessWorker] is_pixi=True, python=/tmp/comfytest-home/.ce/.pixi/envs/sharp-nodes/bin/python\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m got prompt\n  [ComfyUI] [SubprocessWorker] launching cmd=['/tmp/comfytest-home/.pixi/bin/pixi', 'run', '--as-is', '--manifest-path', '/tmp/comfytest-home/.ce/pixi.toml', '-e', 'sharp-nodes', 'python', '/tmp/comfyui_pvenv_qlpb5zhh/persistent_worker.py', 'abstract://comfy_worker_8bd7d9f62285']...\n  [ComfyUI] [comfy-env] Running LoadSharpModel: precision: 'auto'\n  [ComfyUI] [SubprocessWorker] socket_addr=abstract://comfy_worker_8bd7d9f62285 (abstract, no filesystem path)\n  [ComfyUI] [VRAM] Before LoadSharpModel: 4817 / 24122 MB\n  [ComfyUI] [SubprocessWorker] call_method: nodes.load_model.LoadSharpModel.EXECUTE_NORMALIZED\n  [ComfyUI] [SubprocessWorker] acquired lock, ensuring started...\n  [ComfyUI] [ComfyUI-Sharp] Health check: ping (timeout=5.0s)...\n  [ComfyUI] [ComfyUI-Sharp] Health check: failed\n  [ComfyUI] [ComfyUI-Sharp] Socket unhealthy, restarting worker...\n  [ComfyUI] [worker:ComfyUI-Sharp] ERROR: Failed to import comfy_kitchen, Error: No module named 'comfy_kitchen', fp8 and fp4 support will not be available.\n  [ComfyUI] [worker:ComfyUI-Sharp] WARNING: comfy_kitchen does not support stochastic FP8 rounding, please update comfy_kitchen.\n  [ComfyUI] [SubprocessWorker] worker started/confirmed\n  [ComfyUI] [SubprocessWorker] serializing kwargs to shm...\n  [ComfyUI] [SubprocessWorker] created 0 shm blocks\n  [ComfyUI] [SubprocessWorker] sending request via socket...\n  [ComfyUI] [SubprocessWorker] call_id=2 sending call_method\n  [ComfyUI] [worker:ComfyUI-Sharp] [LoadSharpModel] precision='auto' -> dtype=bf16 (device=cuda:0)\n  [ComfyUI] [worker:ComfyUI-Sharp] [LoadSharpModel] checkpoint already present, reusing (no download): /workspaces/Sharp-1914/ComfyUI/models/sharp/sharp_2572gikvuh.pt (2679.6 MB)\n  [ComfyUI] [worker:ComfyUI-Sharp] [LoadSharpModel] checkpoint ready: /workspaces/Sharp-1914/ComfyUI/models/sharp/sharp_2572gikvuh.pt (2679.6 MB)\n  [ComfyUI] [worker:ComfyUI-Sharp] [LoadSharpModel] config: precision=auto -> dtype=bf16\n  [ComfyUI] [SubprocessWorker] got response: ok\n  [ComfyUI] [comfy-env] [VRAM] Before offload 'MoGeWrapper_1' (631 MB) -> cpu: 4817 / 24122 MB\n  [ComfyUI] [ComfyUI-MoGe2] Health check: ping (timeout=5.0s)...\n  [ComfyUI] [ComfyUI-MoGe2] Health check: ok\n  [ComfyUI] [SubprocessWorker] call_id=4 sending model_to_device\n  [ComfyUI] \u001b[1m\u001b[33m[WARNING]\u001b[0m Model 'MoGeWrapper_1' gone from worker (restarted?), treating as offloaded\n  [ComfyUI] [comfy-env] [VRAM] After offload 'MoGeWrapper_1' (631 MB): 4817 / 24122 MB\n  [ComfyUI] [comfy-env] LoadSharpModel done (1.64s): result: NodeOutput\n  [ComfyUI] [VRAM] After LoadSharpModel: 4817 / 24122 MB\n  [ComfyUI] [comfy-env] Running SharpPredict: focal_length_mm: 30.0, output_prefix: 'sharp', model: dict, image: torch.float32 [1,1024,1536,3] cpu\n  [ComfyUI] [VRAM] Before SharpPredict: 4817 / 24122 MB\n  [ComfyUI] [SubprocessWorker] call_method: nodes.predict.SharpPredict.EXECUTE_NORMALIZED\n  [ComfyUI] [SubprocessWorker] acquired lock, ensuring started...\n  [ComfyUI] [ComfyUI-Sharp] Health check: ping (timeout=5.0s)...\n  [ComfyUI] [ComfyUI-Sharp] Health check: ok\n  [ComfyUI] [SubprocessWorker] worker started/confirmed\n  [ComfyUI] [comfy-env] PRE-SERIALIZE 'image' shape: torch.Size([1, 1024, 1536, 3])\n  [ComfyUI] [SubprocessWorker] serializing kwargs to shm...\n  [ComfyUI] [SubprocessWorker] created 0 shm blocks\n  [ComfyUI] [SubprocessWorker] sending request via socket...\n  [ComfyUI] [SubprocessWorker] call_id=3 sending call_method\n  [ComfyUI] [worker:ComfyUI-Sharp] [comfy-env] _from_shm got dict with keys: ['__type__', 'strategy', 'parent_pid', 'fd', 'storage_size']\n  [ComfyUI] [worker:ComfyUI-Sharp] [comfy-env] DESERIALIZE TensorRef: tensor_size=[1, 1024, 1536, 3]\n  [ComfyUI] [worker:ComfyUI-Sharp] [comfy-env] DESERIALIZED tensor shape: torch.Size([1, 1024, 1536, 3])\n  Frame 2 saved (frame_001.jpg, t=2.3s)\n  [capture-periodic] freeze=1ms shot=108ms unfreeze=3ms saved=True\n  Frame 3 saved (frame_002.jpg, t=4.5s)\n  [capture-periodic] freeze=1ms shot=108ms unfreeze=3ms saved=True\n  [capture-loop] iter=50 t=5.6s state={'complete': False, 'error': None, 'executedCount': 0, 'wsState': 1} eval_ms=1 ram=10.8/31.3GB (34%) swap=1.8/32.0GB (6%) cpu=15% browser_rss=1.19GB\n  [ComfyUI] [worker:ComfyUI-Sharp] [LoadSharpModel] loading checkpoint from /workspaces/Sharp-1914/ComfyUI/models/sharp/sharp_2572gikvuh.pt (dtype=bf16)\n  [ComfyUI] [worker:ComfyUI-Sharp] [LoadSharpModel] initializing model on meta device...\n  [ComfyUI] [worker:ComfyUI-Sharp] [LoadSharpModel] model ready (torch.bfloat16)\n  [ComfyUI] [SubprocessWorker] callback 'request_vram_budget' from call_id=3\n  [ComfyUI] [comfy-env] VRAM request: 2.81GB, free before eviction: 20.24GB\n  [ComfyUI] [comfy-env] VRAM after eviction: 20.24GB\n  [ComfyUI] [worker:ComfyUI-Sharp] [VRAM] +715MB (now 715MB) reserved=736MB peak=715MB\n  [ComfyUI] [worker:ComfyUI-Sharp] [VRAM] +2563MB (now 3278MB) reserved=3456MB peak=3350MB\n  [ComfyUI] [worker:ComfyUI-Sharp] [VRAM] +415MB (now 3693MB) reserved=4032MB peak=3693MB\n  [capture-loop] iter=100 t=11.0s state={'complete': False, 'error': None, 'executedCount': 0, 'wsState': 1} eval_ms=2 ram=11.1/31.3GB (36%) swap=1.8/32.0GB (6%) cpu=13% browser_rss=1.09GB\n  [ComfyUI] [worker:ComfyUI-Sharp] [VRAM] +470MB (now 4164MB) reserved=7840MB peak=7053MB\n  [ComfyUI] [worker:ComfyUI-Sharp] [VRAM] +234MB (now 4398MB) reserved=7360MB peak=7053MB\n  [ComfyUI] [worker:ComfyUI-Sharp] [VRAM] -1544MB (now 2854MB) reserved=7360MB peak=7053MB\n  [ComfyUI] [SubprocessWorker] callback 'report_progress' from call_id=3\n  [ComfyUI] [SubprocessWorker] got response: ok\n  [ComfyUI] [comfy-env] SubprocessWorker: /tmp/comfytest-home/.ce/.pixi/envs/gaussianpack-nodes/bin/python\n  [ComfyUI] [comfy-env] env_vars: {'COMFYUI_BASE': '/workspaces/Sharp-1914/ComfyUI'}\n  [ComfyUI] [comfy-env] Created 1 model patchers: ['RGBGaussianPredictor_1']\n  [ComfyUI] [ComfyUI-GaussianPack] python: /tmp/comfytest-home/.ce/.pixi/envs/gaussianpack-nodes/bin/python\n  [ComfyUI] [comfy-env] SharpPredict done (12.18s): result: NodeOutput\n  [ComfyUI] [VRAM] After SharpPredict: 12291 / 24122 MB\n  [ComfyUI] [ComfyUI-GaussianPack] sys_path sent to worker: ['/workspaces/Sharp-1914/ComfyUI', '/workspaces/Sharp-1914/ComfyUI/custom_nodes/ComfyUI-GaussianPack', '/tmp/comfytest-home/.ce/.pixi/envs/gaussianpack-nodes/lib/python3.1/site-packages', '/workspaces/Sharp-1914/ComfyUI/custom_nodes/ComfyUI-GaussianPack']\n  [ComfyUI] [VRAM] Loaded: SubprocessModel (2679 MB)\n  [ComfyUI] [SubprocessWorker] is_pixi=True, python=/tmp/comfytest-home/.ce/.pixi/envs/gaussianpack-nodes/bin/python\n  [ComfyUI] [comfy-env] Running PreviewGaussianSharp: ply_path: '/workspaces/Sharp-1914/ComfyUI/output/sharp_178086030688..., extrinsics: list[4], intrinsics: list[3]\n  [ComfyUI] [SubprocessWorker] launching cmd=['/tmp/comfytest-home/.pixi/bin/pixi', 'run', '--as-is', '--manifest-path', '/tmp/comfytest-home/.ce/pixi.toml', '-e', 'gaussianpack-nodes', 'python', '/tmp/comfyui_pvenv_ii0358nh/persistent_worker.py', 'abstract://comfy_worker_26954ef0c0d3']...\n  [ComfyUI] [VRAM] Before PreviewGaussianSharp: 12291 / 24122 MB\n  [ComfyUI] [SubprocessWorker] socket_addr=abstract://comfy_worker_26954ef0c0d3 (abstract, no filesystem path)\n  [ComfyUI] [VRAM] Loaded: SubprocessModel (2679 MB)\n  [ComfyUI] [SubprocessWorker] call_method: nodes.preview_gaussian_sharp.PreviewGaussianSharp.EXECUTE_NORMALIZED\n  [ComfyUI] [SubprocessWorker] acquired lock, ensuring started...\n  [ComfyUI] [SubprocessWorker] worker started/confirmed\n  [ComfyUI] [SubprocessWorker] serializing kwargs to shm...\n  [ComfyUI] [SubprocessWorker] created 0 shm blocks\n  [ComfyUI] [SubprocessWorker] sending request via socket...\n  [ComfyUI] [SubprocessWorker] call_id=1 sending call_method\n  Frame 4 saved (frame_003.jpg, t=15.4s)\n  [capture-periodic] freeze=1ms shot=96ms unfreeze=2ms saved=True\n  [ComfyUI] [worker:ComfyUI-GaussianPack] WARNING: SPZ route not registered (ModuleNotFoundError: No module named 'comfy_aimdo')\n  [ComfyUI] [SubprocessWorker] got response: ok\n  [ComfyUI] [comfy-env] PreviewGaussianSharp done (1.53s): result: NodeOutput\n  [ComfyUI] [VRAM] After PreviewGaussianSharp: 12291 / 24122 MB\n  [ComfyUI] [VRAM] Loaded: SubprocessModel (2679 MB)\n  [ComfyUI] \u001b[32m[INFO]\u001b[0m \u001b[32mPrompt executed in 15.38 seconds\u001b[0m\n  [ComfyUI] [comfy-env] [VRAM] Before offload 'RGBGaussianPredictor_1' (2679 MB) -> cpu: 12291 / 24122 MB\n  [ComfyUI] [ComfyUI-Sharp] Health check: ping (timeout=5.0s)...\n  [ComfyUI] [ComfyUI-Sharp] Health check: ok\n  [ComfyUI] [SubprocessWorker] call_id=4 sending model_to_device\n  [ComfyUI] [worker:ComfyUI-Sharp] [VRAM] -360MB (now 2495MB) reserved=7360MB peak=7053MB"
    },
    {
      "file": "frame_005.jpg",
      "time": 50.60527849197388,
      "log": "Final screenshot"
    }
  ],
  "total_time": 48.25
}