[00:00] Crash dump logging enabled: /logs/DepthAnythingV3-0759/dev/linux-gpu/crash_dump.log [00:00] [00:00] [1/5] SYNTAX [00:00] ---------------------------------------- [00:00] [DEBUG] server=None, server_url=None, api=None [00:00] Found pyproject.toml (modern format) [00:00] Found requirements.txt (legacy format) [00:00] Unicode check: OK (all characters cp1252-safe) [00:00] Forbidden patterns check: OK [00:00] [SYNTAX] PASSED [00:00] [00:00] [2/5] INSTALL [00:00] ---------------------------------------- [00:00] [DEBUG] server=None, server_url=None, api=None [00:00] Setting up ComfyUI... [00:00] Creating virtual environment... [00:00] Running: uv venv /workspaces/DepthAnythingV3-0759/.venv --python 3.12 [00:00] Cloning ComfyUI (latest)... [00:00] Running: git clone --depth 1 https://github.com/comfyanonymous/ComfyUI.git /workspaces/DepthAnythingV3-0759/ComfyUI [00:02] Installing ComfyUI requirements... [00:02] Running: uv pip install --python /workspaces/DepthAnythingV3-0759/.venv/bin/python --index-url https://download.pytorch.org/whl/cu128 --extra-index-url https://pypi.org/simple --index-strategy unsafe-best-match -r /workspaces/DepthAnythingV3-0759/ComfyUI/requirements.txt [01:20] Installing custom node... [01:20] Copying ComfyUI-DepthAnythingV3 to custom_nodes/... [01:20] Installing node requirements... [01:20] Running: uv pip install --python /workspaces/DepthAnythingV3-0759/.venv/bin/python --index-url https://download.pytorch.org/whl/cu128 --extra-index-url https://pypi.org/simple --index-strategy unsafe-best-match -r /workspaces/DepthAnythingV3-0759/ComfyUI/custom_nodes/ComfyUI-DepthAnythingV3/requirements.txt [01:23] Running install.py... [01:23] Running: /workspaces/DepthAnythingV3-0759/.venv/bin/python /workspaces/DepthAnythingV3-0759/ComfyUI/custom_nodes/ComfyUI-DepthAnythingV3/install.py [01:24] [01:24] Installing 3 node dependencies... [01:24] Cloning ComfyUI-VideoHelperSuite... [01:24] Installing requirements for ComfyUI-VideoHelperSuite... [01:26] Cloning ComfyUI-KJNodes... [01:29] Installing requirements for ComfyUI-KJNodes... [01:31] Cloning ComfyUI-GeometryPack... [01:36] Installing requirements for ComfyUI-GeometryPack... [01:37] Running install.py for ComfyUI-GeometryPack... [01:39] [comfy-env] Workspace: /workspaces/DepthAnythingV3-0759/ComfyUI/.ce [01:39] [comfy-env] ComfyUI: /workspaces/DepthAnythingV3-0759/ComfyUI [01:39] [comfy-env] Found 2 node config(s): [01:39] - depthanythingv3-nodes <- custom_nodes/ComfyUI-DepthAnythingV3/nodes/comfy-env.toml (python=host) [01:39] - geometrypack-nodes <- custom_nodes/ComfyUI-GeometryPack/nodes/comfy-env.toml (python=host) [01:39] [comfy-env] Bootstrap interpreter has python 3.12 + torch 2.11.0 (cu128). Looking for cuda-only wheels matching cu128/torch2.11/cp312. [01:39] [comfy-env] GPU: Detection: smi [01:39] Driver: 580.126.20 [01:39] CUDA: 13.0 [01:39] Recommended: CUDA 13.0 [01:39] [01:39] GPU 0: NVIDIA GeForce RTX 3090 (sm_86) [Ampere] 24576MB [01:39] [comfy-env] Comfyui feature -> torch 2.11.0 from https://download.pytorch.org/whl/cu128 (CUDA 12.8 via bootstrap torch) [01:39] [comfy-env] cuda-wheels: 3 package(s) need a matched (cuda, torch, python) combo: ['flash_attn', 'sageattention', 'cumesh'] [01:39] [comfy-env] cuda-wheels tier 1 (bootstrap): probing cu12.8/torch2.11/cp312 [01:39] [cuda-wheels] Looking up flash_attn: cu128 torch2.11 cp312 manylinux linux [01:39] [cuda-wheels] https://pozzettiandrea.github.io/cuda-wheels/v2/flash_attn/: HTTPError: HTTP Error 404: Not Found [01:39] [cuda-wheels] Found: flash_attn-2.8.3+cu128torch2.11-cp312-cp312-manylinux_2_34_x86_64.manylinux_2_35_x86_64.whl [01:39] [cuda-wheels] Looking up sageattention: cu128 torch2.11 cp312 manylinux linux [01:40] [cuda-wheels] Found: sageattention-2.2.0+cu128torch2.11-cp312-cp312-manylinux_2_34_x86_64.manylinux_2_35_x86_64.whl [01:40] [cuda-wheels] Looking up cumesh: cu128 torch2.11 cp312 manylinux linux [01:40] [cuda-wheels] Found: cumesh-0.0.1+cu128torch2.11-cp312-cp312-manylinux_2_35_x86_64.whl [01:40] [comfy-env] cuda-wheels combo: cu12.8/torch2.11 (bootstrap matches; per-node envs will pin to this) [01:40] [comfy-env] Root conda deps from ComfyUI-DepthAnythingV3: ['av', 'ffmpeg'] [01:40] [comfy-env] Root conda deps from ComfyUI-GeometryPack: ['av', 'ffmpeg'] [01:40] [comfy-env] Comfyui feature pin: torch ==2.11.0, torchvision ==0.26.*, torchaudio ==2.11.* [01:40] [comfy-env] Host glibc 2.35 -> comfyui feature system-requirements [01:40] [comfy-env] Root conda deps: ['av', 'ffmpeg'] [01:40] [comfy-env] comfyui312: python 3.12, torch from https://download.pytorch.org/whl/cu128 [01:40] [comfy-env] cuda-wheels combo: per-node cuda features will pin ['torch', 'torchaudio', 'torchvision'] via pypi-options.dependency-overrides (https://download.pytorch.org/whl/cu128) [01:40] [cuda-wheels] Looking up flash_attn: cu128 torch2.11 cp312 manylinux linux [01:40] [cuda-wheels] https://pozzettiandrea.github.io/cuda-wheels/v2/flash_attn/: HTTPError: HTTP Error 404: Not Found [01:40] [cuda-wheels] Found: flash_attn-2.8.3+cu128torch2.11-cp312-cp312-manylinux_2_34_x86_64.manylinux_2_35_x86_64.whl [01:40] [cuda-wheels] Looking up sageattention: cu128 torch2.11 cp312 manylinux linux [01:40] [cuda-wheels] Found: sageattention-2.2.0+cu128torch2.11-cp312-cp312-manylinux_2_34_x86_64.manylinux_2_35_x86_64.whl [01:40] [comfy-env] depthanythingv3-nodes: cuda-wheels deferred for post-pixi install (flash_attn, sageattention) [01:40] [comfy-env] depthanythingv3-nodes: torch override -> torch==2.11.0, torchvision==0.26.*, torchaudio==2.11.* from https://download.pytorch.org/whl/cu128 [01:40] [cuda-wheels] Looking up cumesh: cu128 torch2.11 cp312 manylinux linux [01:40] [cuda-wheels] Found: cumesh-0.0.1+cu128torch2.11-cp312-cp312-manylinux_2_35_x86_64.whl [01:40] [comfy-env] geometrypack-nodes: cuda-wheels deferred for post-pixi install (cumesh) [01:40] [comfy-env] geometrypack-nodes: torch override -> torch==2.11.0, torchvision==0.26.*, torchaudio==2.11.* from https://download.pytorch.org/whl/cu128 [01:40] Generated /workspaces/DepthAnythingV3-0759/ComfyUI/.ce/pixi.toml [01:40] [comfy-env] Installing 2 environment(s): [01:40] - depthanythingv3-nodes (python=host, 9 pypi deps, 2 cuda wheels) [01:40] - geometrypack-nodes (python=host, 13 pypi deps, 1 cuda wheels) [01:40] [comfy-env] Running `pixi install --all`... [02:06] WARN Skipped running the post-link scripts because `run-post-link-scripts` = `false` [02:06] - bin/.librsvg-pre-unlink.sh [02:06] To enable them, run: [02:06] pixi config set --local run-post-link-scripts insecure [02:06] More info: [02:06] https://pixi.sh/latest/reference/pixi_configuration/#run-post-link-scripts [02:08] WARN Skipped running the post-link scripts because `run-post-link-scripts` = `false` [02:08] - bin/.librsvg-pre-unlink.sh [02:08] To enable them, run: [02:08] pixi config set --local run-post-link-scripts insecure [02:08] More info: [02:08] https://pixi.sh/latest/reference/pixi_configuration/#run-post-link-scripts [02:09] WARN Skipped running the post-link scripts because `run-post-link-scripts` = `false` [02:09] - bin/.librsvg-pre-unlink.sh [02:09] To enable them, run: [02:09] pixi config set --local run-post-link-scripts insecure [02:09] More info: [02:09] https://pixi.sh/latest/reference/pixi_configuration/#run-post-link-scripts [03:26] The following environments have been installed: default, comfyui, depthanythingv3-nodes, geometrypack-nodes. [03:26] [comfy-env] Note: env `default` is on disk but no node declares it in this run. Leaving as-is. Remove via `pixi clean --environment default` if intended. [03:26] [comfy-env] depthanythingv3-nodes: installing cuda-wheels with --no-deps (flash_attn, sageattention) [03:26] Using Python 3.12.13 environment at: .pixi/envs/depthanythingv3-nodes [03:28] Resolved 2 packages in 2.14s [03:28] Downloading flash-attn (242.3MiB) [03:28] Downloading sageattention (53.2MiB) [03:30] Downloaded sageattention [03:35] Downloaded flash-attn [03:35] Prepared 2 packages in 6.78s [03:35] Installed 2 packages in 10ms [03:35] + flash-attn==2.8.3+cu128torch2.11 (from https://github.com/PozzettiAndrea/cuda-wheels/releases/download/flash_attn-latest/flash_attn-2.8.3%2Bcu128torch2.11-cp312-cp312-manylinux_2_34_x86_64.manylinux_2_35_x86_64.whl) [03:35] + sageattention==2.2.0+cu128torch2.11 (from https://github.com/PozzettiAndrea/cuda-wheels/releases/download/sageattention-latest/sageattention-2.2.0%2Bcu128torch2.11-cp312-cp312-manylinux_2_34_x86_64.manylinux_2_35_x86_64.whl) [03:35] [comfy-env] geometrypack-nodes: installing cuda-wheels with --no-deps (cumesh) [03:35] Using Python 3.12.13 environment at: .pixi/envs/geometrypack-nodes [03:36] Resolved 1 package in 1.10s [03:36] Downloading cumesh (19.5MiB) [03:37] Downloaded cumesh [03:37] Prepared 1 package in 321ms [03:37] Installed 1 package in 8ms [03:37] + cumesh==0.0.1+cu128torch2.11 (from https://github.com/PozzettiAndrea/cuda-wheels/releases/download/cumesh-latest/cumesh-0.0.1%2Bcu128torch2.11-cp312-cp312-manylinux_2_35_x86_64.whl) [03:37] [comfy-env] Install log: /workspaces/DepthAnythingV3-0759/ComfyUI/.ce/install.log [03:37] [03:37] Installation complete! [03:37] [03:37] [requirements] Re-installing main package requirements... [03:37] Installing requirements for ComfyUI-DepthAnythingV3... [03:37] [comfy-env] Workspace: /workspaces/DepthAnythingV3-0759/ComfyUI/.ce [03:37] [comfy-env] ComfyUI: /workspaces/DepthAnythingV3-0759/ComfyUI [03:37] [comfy-env] Found 2 node config(s): [03:37] - depthanythingv3-nodes <- custom_nodes/ComfyUI-DepthAnythingV3/nodes/comfy-env.toml (python=host) [03:37] - geometrypack-nodes <- custom_nodes/ComfyUI-GeometryPack/nodes/comfy-env.toml (python=host) [03:37] [comfy-env] Bootstrap interpreter has python 3.12 + torch 2.11.0 (cu128). Looking for cuda-only wheels matching cu128/torch2.11/cp312. [03:37] [comfy-env] GPU: Detection: smi [03:37] Driver: 580.126.20 [03:37] CUDA: 13.0 [03:37] Recommended: CUDA 13.0 [03:37] [03:37] GPU 0: NVIDIA GeForce RTX 3090 (sm_86) [Ampere] 24576MB [03:37] [comfy-env] Comfyui feature -> torch 2.11.0 from https://download.pytorch.org/whl/cu128 (CUDA 12.8 via bootstrap torch) [03:37] [comfy-env] cuda-wheels: 3 package(s) need a matched (cuda, torch, python) combo: ['flash_attn', 'sageattention', 'cumesh'] [03:37] [comfy-env] cuda-wheels tier 1 (bootstrap): probing cu12.8/torch2.11/cp312 [03:37] [cuda-wheels] Looking up flash_attn: cu128 torch2.11 cp312 manylinux linux [03:37] [cuda-wheels] https://pozzettiandrea.github.io/cuda-wheels/v2/flash_attn/: HTTPError: HTTP Error 404: Not Found [03:37] [cuda-wheels] Found: flash_attn-2.8.3+cu128torch2.11-cp312-cp312-manylinux_2_34_x86_64.manylinux_2_35_x86_64.whl [03:37] [cuda-wheels] Looking up sageattention: cu128 torch2.11 cp312 manylinux linux [03:38] [cuda-wheels] Found: sageattention-2.2.0+cu128torch2.11-cp312-cp312-manylinux_2_34_x86_64.manylinux_2_35_x86_64.whl [03:38] [cuda-wheels] Looking up cumesh: cu128 torch2.11 cp312 manylinux linux [03:38] [cuda-wheels] Found: cumesh-0.0.1+cu128torch2.11-cp312-cp312-manylinux_2_35_x86_64.whl [03:38] [comfy-env] cuda-wheels combo: cu12.8/torch2.11 (bootstrap matches; per-node envs will pin to this) [03:38] [comfy-env] Root conda deps from ComfyUI-DepthAnythingV3: ['av', 'ffmpeg'] [03:38] [comfy-env] Root conda deps from ComfyUI-GeometryPack: ['av', 'ffmpeg'] [03:38] [comfy-env] Comfyui feature pin: torch ==2.11.0, torchvision ==0.26.*, torchaudio ==2.11.* [03:38] [comfy-env] Host glibc 2.35 -> comfyui feature system-requirements [03:38] [comfy-env] Root conda deps: ['av', 'ffmpeg'] [03:38] [comfy-env] comfyui312: python 3.12, torch from https://download.pytorch.org/whl/cu128 [03:38] [comfy-env] cuda-wheels combo: per-node cuda features will pin ['torch', 'torchaudio', 'torchvision'] via pypi-options.dependency-overrides (https://download.pytorch.org/whl/cu128) [03:38] [cuda-wheels] Looking up flash_attn: cu128 torch2.11 cp312 manylinux linux [03:38] [cuda-wheels] https://pozzettiandrea.github.io/cuda-wheels/v2/flash_attn/: HTTPError: HTTP Error 404: Not Found [03:38] [cuda-wheels] Found: flash_attn-2.8.3+cu128torch2.11-cp312-cp312-manylinux_2_34_x86_64.manylinux_2_35_x86_64.whl [03:38] [cuda-wheels] Looking up sageattention: cu128 torch2.11 cp312 manylinux linux [03:38] [cuda-wheels] Found: sageattention-2.2.0+cu128torch2.11-cp312-cp312-manylinux_2_34_x86_64.manylinux_2_35_x86_64.whl [03:38] [comfy-env] depthanythingv3-nodes: cuda-wheels deferred for post-pixi install (flash_attn, sageattention) [03:38] [comfy-env] depthanythingv3-nodes: torch override -> torch==2.11.0, torchvision==0.26.*, torchaudio==2.11.* from https://download.pytorch.org/whl/cu128 [03:38] [cuda-wheels] Looking up cumesh: cu128 torch2.11 cp312 manylinux linux [03:38] [cuda-wheels] Found: cumesh-0.0.1+cu128torch2.11-cp312-cp312-manylinux_2_35_x86_64.whl [03:38] [comfy-env] geometrypack-nodes: cuda-wheels deferred for post-pixi install (cumesh) [03:38] [comfy-env] geometrypack-nodes: torch override -> torch==2.11.0, torchvision==0.26.*, torchaudio==2.11.* from https://download.pytorch.org/whl/cu128 [03:38] Generated /workspaces/DepthAnythingV3-0759/ComfyUI/.ce/pixi.toml [03:38] [comfy-env] Installing 2 environment(s): [03:38] - depthanythingv3-nodes (python=host, 9 pypi deps, 2 cuda wheels) [03:38] - geometrypack-nodes (python=host, 13 pypi deps, 1 cuda wheels) [03:38] [comfy-env] Running `pixi install --all`... [03:38] The following environments have been installed: default, comfyui, depthanythingv3-nodes, geometrypack-nodes. [03:38] [comfy-env] Note: env `default` is on disk but no node declares it in this run. Leaving as-is. Remove via `pixi clean --environment default` if intended. [03:38] [comfy-env] depthanythingv3-nodes: installing cuda-wheels with --no-deps (flash_attn, sageattention) [03:38] Using Python 3.12.13 environment at: .pixi/envs/depthanythingv3-nodes [03:38] Checked 2 packages in 69ms [03:38] [comfy-env] geometrypack-nodes: installing cuda-wheels with --no-deps (cumesh) [03:38] Using Python 3.12.13 environment at: .pixi/envs/geometrypack-nodes [03:38] Checked 1 package in 67ms [03:38] [comfy-env] Install log: /workspaces/DepthAnythingV3-0759/ComfyUI/.ce/install.log [03:38] [03:38] Installation complete! [03:38] Installing 3 node dependency(ies)... [03:38] ComfyUI-VideoHelperSuite from Kosinkadink/ComfyUI-VideoHelperSuite [03:38] ComfyUI-VideoHelperSuite already exists, skipping... [03:38] ComfyUI-KJNodes from kijai/ComfyUI-KJNodes [03:38] ComfyUI-KJNodes already exists, skipping... [03:38] ComfyUI-GeometryPack from PozzettiAndrea/ComfyUI-GeometryPack [03:38] ComfyUI-GeometryPack already exists, skipping... [03:38] Installing validation endpoint... [03:38] Cloning PozzettiAndrea/ComfyUI-validate-endpoint... [03:38] Running: git clone --depth 1 https://github.com/PozzettiAndrea/ComfyUI-validate-endpoint.git /workspaces/DepthAnythingV3-0759/ComfyUI/custom_nodes/ComfyUI-validate-endpoint [03:39] CUDA packages installed (no mock): flash_attn, sageattention [03:39] [INSTALL] PASSED [03:39] [03:39] [3/5] REGISTRATION [03:39] ---------------------------------------- [03:39] [DEBUG] server=None, server_url=None, api=None [03:39] Starting ComfyUI server... [03:39] Starting ComfyUI server on port 41889... [03:39] Starting ComfyUI server on port 41889... [03:39] Waiting for server to be ready (timeout: 600s)... [03:41] [ComfyUI] setup plugin alembic.autogenerate.schemas [03:41] [ComfyUI] setup plugin alembic.autogenerate.tables [03:41] [ComfyUI] setup plugin alembic.autogenerate.types [03:41] [ComfyUI] setup plugin alembic.autogenerate.constraints [03:41] [ComfyUI] setup plugin alembic.autogenerate.defaults [03:41] [ComfyUI] setup plugin alembic.autogenerate.comments [03:41] [ComfyUI] [comfy-env] ComfyUI-DepthAnythingV3: 1 isolation env(s): [03:41] [ComfyUI] [comfy-env] nodes -> /workspaces/DepthAnythingV3-0759/ComfyUI/custom_nodes/ComfyUI-DepthAnythingV3/nodes [03:41] [ComfyUI] [comfy-env] prestartup complete [03:41] [ComfyUI] [comfy-env] ComfyUI-GeometryPack: 1 isolation env(s): [03:41] [ComfyUI] [comfy-env] nodes -> /workspaces/DepthAnythingV3-0759/ComfyUI/custom_nodes/ComfyUI-GeometryPack/nodes [03:41] [ComfyUI] [comfy-env] prestartup complete [03:41] [ComfyUI] [03:41] [ComfyUI] Prestartup times for custom nodes: [03:41] [ComfyUI] 0.1 seconds: /workspaces/DepthAnythingV3-0759/ComfyUI/custom_nodes/ComfyUI-GeometryPack [03:41] [ComfyUI] 0.2 seconds: /workspaces/DepthAnythingV3-0759/ComfyUI/custom_nodes/ComfyUI-DepthAnythingV3 [03:41] [ComfyUI] [03:48] [ComfyUI] WARNING: You need pytorch with cu130 or higher to use optimized CUDA operations. [03:48] [ComfyUI] Found comfy_kitchen backend triton: {'available': True, 'disabled': True, 'unavailable_reason': None, 'capabilities': ['apply_rope', 'apply_rope1', 'dequantize_nvfp4', 'dequantize_per_tensor_fp8', 'quantize_mxfp8', 'quantize_nvfp4', 'quantize_per_tensor_fp8']} [03:48] [ComfyUI] Found comfy_kitchen backend eager: {'available': True, 'disabled': False, 'unavailable_reason': None, 'capabilities': ['apply_rope', 'apply_rope1', 'dequantize_mxfp8', 'dequantize_nvfp4', 'dequantize_per_tensor_fp8', 'quantize_mxfp8', 'quantize_nvfp4', 'quantize_per_tensor_fp8', 'scaled_mm_mxfp8', 'scaled_mm_nvfp4']} [03:48] [ComfyUI] Found comfy_kitchen backend cuda: {'available': True, 'disabled': True, 'unavailable_reason': None, 'capabilities': ['apply_rope', 'apply_rope1', 'dequantize_nvfp4', 'dequantize_per_tensor_fp8', 'quantize_mxfp8', 'quantize_nvfp4', 'quantize_per_tensor_fp8']} [03:48] [ComfyUI] Checkpoint files will always be loaded safely. [03:48] [ComfyUI] Total VRAM 24124 MB, total RAM 31924 MB [03:48] [ComfyUI] pytorch version: 2.11.0+cu128 [03:48] [ComfyUI] Set vram state to: NORMAL_VRAM [03:48] [ComfyUI] Device: cuda:0 NVIDIA GeForce RTX 3090 : cudaMallocAsync [03:48] [ComfyUI] Using async weight offloading with 2 streams [03:48] [ComfyUI] Enabled pinned memory 28731.0 [03:49] [ComfyUI] Using pytorch attention [03:54] [ComfyUI] aimdo: /project/src-posix/cuda-funchooks.c:52:DEBUG:aimdo_setup_hooks: hooks successfully installed [03:54] [ComfyUI] aimdo: /project/src/control.c:151:INFO:comfy-aimdo inited for GPU: NVIDIA GeForce RTX 3090 (VRAM: 24124 MB) [03:54] [ComfyUI] DynamicVRAM support detected and enabled [03:54] [ComfyUI] Python version: 3.12.13 (main, May 4 2026, 21:09:48) [Clang 22.1.3 ] [03:54] [ComfyUI] ComfyUI version: 0.20.1 [03:54] [ComfyUI] comfy-aimdo version: 0.3.0 [03:54] [ComfyUI] comfy-kitchen version: 0.2.8 [03:54] [ComfyUI] ****** User settings have been changed to be stored on the server instead of browser storage. ****** [03:54] [ComfyUI] ****** For multi-user setups add the --multi-user CLI argument to enable multiple user profiles. ****** [03:54] [ComfyUI] ComfyUI frontend version: 1.43.18 [03:54] [ComfyUI] [Prompt Server] web root: /workspaces/DepthAnythingV3-0759/.venv/lib/python3.12/site-packages/comfyui_frontend_package/static [03:54] [ComfyUI] Asset seeder disabled [03:57] [ComfyUI] [comfy-env] Version: 0.3.78 [03:57] [ComfyUI] [comfy-env] register_nodes: ComfyUI-DepthAnythingV3 [03:57] [ComfyUI] [comfy-env] ComfyUI source dir: /workspaces/DepthAnythingV3-0759/ComfyUI [03:57] [ComfyUI] [comfy-env] Found 1 isolation env(s) [03:57] [ComfyUI] [comfy-env] Scanning nodes metadata (isolation root)... [03:57] [ComfyUI] [comfy-env] Metadata scan: /root/.pixi/bin/pixi run --as-is --manifest-path /workspaces/DepthAnythingV3-0759/ComfyUI/.ce/pixi.toml -e depthanythingv3-nodes python /tmp/comfy_meta_6empzund.py /workspaces/DepthAnythingV3-0759/ComfyUI/custom_nodes/ComfyUI-DepthAnythingV3 nodes [03:57] [ComfyUI] [comfy-env] Scan env PATH for nodes: [03:57] [ComfyUI] [comfy-env] [0] /root/.local/bin [03:57] [ComfyUI] [comfy-env] [1] /usr/local/nvidia/bin [03:57] [ComfyUI] [comfy-env] [2] /usr/local/cuda/bin [03:57] [ComfyUI] [comfy-env] [3] /usr/local/sbin [03:57] [ComfyUI] [comfy-env] [4] /usr/local/bin [03:57] [ComfyUI] [comfy-env] [5] /usr/sbin [03:57] [ComfyUI] [comfy-env] [6] /usr/bin [03:57] [ComfyUI] [comfy-env] [7] /sbin [03:57] [ComfyUI] [comfy-env] [8] /bin [04:04] [ComfyUI] [comfy-env] Metadata scan stderr for nodes: [04:04] [ComfyUI] [comfy-env] [meta-scan] importing nodes from /workspaces/DepthAnythingV3-0759/ComfyUI/custom_nodes/ComfyUI-DepthAnythingV3 [04:04] [ComfyUI] [comfy-env] WARNING:root:WARNING: You need pytorch with cu130 or higher to use optimized CUDA operations. [04:04] [ComfyUI] [comfy-env] [meta-scan] import OK [04:04] [ComfyUI] [comfy-env] Scanned nodes: 15 nodes (7.5s) [04:04] [ComfyUI] [comfy-env] Scanned nodes root: 15 nodes (7.5s) [04:04] [ComfyUI] [comfy-env] Imported nodes root: 15 nodes (isolation) [04:04] [ComfyUI] [comfy-env] Registered 15 total nodes [04:05] [ComfyUI] loading... [04:05] [ComfyUI] calling register_nodes [04:05] [ComfyUI] [comfy-env] Version: 0.3.78 [04:05] [ComfyUI] [comfy-env] register_nodes: ComfyUI-GeometryPack [04:05] [ComfyUI] [comfy-env] ComfyUI source dir: /workspaces/DepthAnythingV3-0759/ComfyUI [04:05] [ComfyUI] [comfy-env] Found 1 isolation env(s) [04:05] [ComfyUI] [comfy-env] Scanning nodes metadata (isolation root)... [04:05] [ComfyUI] [comfy-env] Metadata scan: /root/.pixi/bin/pixi run --as-is --manifest-path /workspaces/DepthAnythingV3-0759/ComfyUI/.ce/pixi.toml -e geometrypack-nodes python /tmp/comfy_meta_kqh8nwa6.py /workspaces/DepthAnythingV3-0759/ComfyUI/custom_nodes/ComfyUI-GeometryPack nodes [04:05] [ComfyUI] [comfy-env] Scan env PATH for nodes: [04:05] [ComfyUI] [comfy-env] [0] /root/.local/bin [04:05] [ComfyUI] [comfy-env] [1] /usr/local/nvidia/bin [04:05] [ComfyUI] [comfy-env] [2] /usr/local/cuda/bin [04:05] [ComfyUI] [comfy-env] [3] /usr/local/sbin [04:05] [ComfyUI] [comfy-env] [4] /usr/local/bin [04:05] [ComfyUI] [comfy-env] [5] /usr/sbin [04:05] [ComfyUI] [comfy-env] [6] /usr/bin [04:05] [ComfyUI] [comfy-env] [7] /sbin [04:05] [ComfyUI] [comfy-env] [8] /bin [04:15] [ComfyUI] [comfy-env] Metadata scan stderr for nodes: [04:15] [ComfyUI] [comfy-env] [meta-scan] importing nodes from /workspaces/DepthAnythingV3-0759/ComfyUI/custom_nodes/ComfyUI-GeometryPack [04:15] [ComfyUI] [comfy-env] WARNING:root:WARNING: You need pytorch with cu130 or higher to use optimized CUDA operations. [04:15] [ComfyUI] [comfy-env] [meta-scan] import OK [04:15] [ComfyUI] [comfy-env] Scanned nodes: 125 nodes (10.6s) [04:15] [ComfyUI] [comfy-env] Scanned nodes root: 125 nodes (10.6s) [04:15] [ComfyUI] [comfy-env] Imported nodes root: 125 nodes (isolation) [04:15] [ComfyUI] [comfy-env] Registered 125 total nodes [04:15] [ComfyUI] [04:15] [ComfyUI] Import times for custom nodes: [04:15] [ComfyUI] 0.0 seconds: /workspaces/DepthAnythingV3-0759/ComfyUI/custom_nodes/websocket_image_save.py [04:15] [ComfyUI] 0.0 seconds: /workspaces/DepthAnythingV3-0759/ComfyUI/custom_nodes/ComfyUI-validate-endpoint [04:15] [ComfyUI] 0.1 seconds: /workspaces/DepthAnythingV3-0759/ComfyUI/custom_nodes/ComfyUI-VideoHelperSuite [04:15] [ComfyUI] 0.3 seconds: /workspaces/DepthAnythingV3-0759/ComfyUI/custom_nodes/ComfyUI-KJNodes [04:15] [ComfyUI] 7.7 seconds: /workspaces/DepthAnythingV3-0759/ComfyUI/custom_nodes/ComfyUI-DepthAnythingV3 [04:15] [ComfyUI] 10.6 seconds: /workspaces/DepthAnythingV3-0759/ComfyUI/custom_nodes/ComfyUI-GeometryPack [04:15] [ComfyUI] [04:15] [ComfyUI] Context impl SQLiteImpl. [04:15] [ComfyUI] Will assume non-transactional DDL. [04:15] [ComfyUI] Context impl SQLiteImpl. [04:15] [ComfyUI] Will assume non-transactional DDL. [04:15] [ComfyUI] Running upgrade -> 0001_assets, Initial assets schema [04:15] [ComfyUI] Revision ID: 0001_assets [04:15] [ComfyUI] Revises: None [04:15] [ComfyUI] Create Date: 2025-12-10 00:00:00 [04:16] [ComfyUI] Running upgrade 0001_assets -> 0002_merge_to_asset_references, Merge AssetInfo and AssetCacheState into unified asset_references table. [04:16] [ComfyUI] Running upgrade 0002_merge_to_asset_references -> 0003_add_metadata_job_id, Add system_metadata and job_id columns to asset_references. [04:16] [ComfyUI] Change preview_id FK from assets.id to asset_references.id. [04:16] [ComfyUI] Database upgraded from None to 0003_add_metadata_job_id [04:16] [ComfyUI] Starting server [04:16] [ComfyUI] [04:16] [ComfyUI] To see the GUI go to: http://127.0.0.1:41889 [04:16] Server responding, waiting for nodes to load... [04:36] Server is ready! [04:36] Checking for import errors in server logs... [04:36] No import errors detected [04:37] Found 1120 registered nodes [04:37] [REGISTRATION] PASSED [04:37] [04:37] [4/5] INSTANTIATION [04:37] ---------------------------------------- [04:37] [DEBUG] server=, server_url=None, api= [04:37] Testing node constructors... [04:37] All 15 node(s) instantiated successfully! [04:37] [INSTANTIATION] PASSED [04:37] [04:37] [5/5] EXECUTION [04:37] ---------------------------------------- [04:37] [DEBUG] server=, server_url=None, api= [04:37] GPU runner - will execute 7 workflow(s) [04:37] Running 7 workflow(s) (all with videos)... [04:37] Installing screenshot dependencies (playwright, pillow)... [04:39] Packages installed, downloading chromium browser... [05:01] Screenshot dependencies installed successfully [05:02] Starting headless browser... [05:02] Installing Vulkan stack (apt)... [05:02] Vulkan install error: [Errno 2] No such file or directory: 'sudo' [05:02] GPU detected (no Vulkan) -- using ANGLE/OpenGL rendering [05:02] Capturing execution frames: advanced.json [05:04] [ComfyUI] [DEPRECATION WARNING] Detected import of deprecated legacy API: /extensions/core/widgetInputs.js. This is likely caused by a custom node extension using outdated APIs. Please update your extensions or contact the extension author for an updated version. [05:10] Frame 1 saved (frame_000.jpg, t=0.0s) [05:10] Validating workflow... [05:10] [ComfyUI] === /validate endpoint called === [05:10] [ComfyUI] Received JSON with keys: ['prompt'] [05:10] Queuing workflow for execution... [05:10] [ComfyUI] Prompt has 11 nodes [05:10] [ComfyUI] Node 2: LoadImage [05:10] [ComfyUI] Inputs: {"image": "bridge.jpg"} [05:10] [ComfyUI] Node 3: DownloadAndLoadDepthAnythingV3Model [05:10] [ComfyUI] Inputs: {"model": "da3_large.safetensors", "precision": "auto", "attention": "auto"} [05:10] [ComfyUI] Node 4: PreviewImage [05:10] [ComfyUI] Inputs: {"images": ["12", 0]} [05:10] [ComfyUI] Node 6: PreviewImage [05:10] [ComfyUI] Inputs: {"images": ["12", 1]} [05:10] [ComfyUI] Node 7: PreviewImage [05:10] [ComfyUI] [comfy-env] SubprocessWorker: /workspaces/DepthAnythingV3-0759/ComfyUI/.ce/.pixi/envs/depthanythingv3-nodes/bin/python [05:10] [ComfyUI] Inputs: {"images": ["12", 2]} [05:10] [ComfyUI] Node 8: PreviewImage [05:10] [ComfyUI] [comfy-env] env_vars: {'COMFYUI_BASE': '/workspaces/DepthAnythingV3-0759/ComfyUI'} [05:10] [ComfyUI] Inputs: {"images": ["12", 3]} [05:10] [ComfyUI] [ComfyUI-DepthAnythingV3] python: /workspaces/DepthAnythingV3-0759/ComfyUI/.ce/.pixi/envs/depthanythingv3-nodes/bin/python [05:10] [ComfyUI] Node 9: PreviewAny [05:10] [ComfyUI] [ComfyUI-DepthAnythingV3] sys_path sent to worker: ['/workspaces/DepthAnythingV3-0759/ComfyUI', '/workspaces/DepthAnythingV3-0759/ComfyUI/custom_nodes/ComfyUI-DepthAnythingV3', '/workspaces/DepthAnythingV3-0759/ComfyUI/.ce/.pixi/envs/depthanythingv3-nodes/lib/python3.1/site-packages', '/workspaces/DepthAnythingV3-0759/ComfyUI/custom_nodes/ComfyUI-DepthAnythingV3'] [05:10] [ComfyUI] Inputs: {"source": ["12", 6]} [05:10] [ComfyUI] [SubprocessWorker] is_pixi=True, python=/workspaces/DepthAnythingV3-0759/ComfyUI/.ce/.pixi/envs/depthanythingv3-nodes/bin/python [05:10] [ComfyUI] Node 10: PreviewAny [05:10] [ComfyUI] [SubprocessWorker] launching cmd=['/root/.pixi/bin/pixi', 'run', '--as-is', '--manifest-path', '/workspaces/DepthAnythingV3-0759/ComfyUI/.ce/pixi.toml', '-e', 'depthanythingv3-nodes', 'python', '/tmp/comfyui_pvenv_h_nkff2u/persistent_worker.py', 'abstract://comfy_worker_a1bd1498d0fe']... [05:10] [ComfyUI] Inputs: {"source": ["12", 5]} [05:10] [ComfyUI] [SubprocessWorker] socket_addr=abstract://comfy_worker_a1bd1498d0fe (abstract, no filesystem path) [05:10] [ComfyUI] Node 11: MaskPreview [05:10] [ComfyUI] Inputs: {"mask": ["12", 7]} [05:10] [ComfyUI] Node 12: DepthAnything_V3 [05:10] [ComfyUI] Inputs: {"normalization_mode": "V2-Style", "resize_method": "resize", "invert_depth": false, "keep_model_size": false, "da3_model": ["3", 0], "images": ["2", 0]} [05:10] [ComfyUI] Node 13: PreviewImage [05:10] [ComfyUI] Inputs: {"images": ["12", 4]} [05:10] [ComfyUI] Generated prompt_id: 561e7bfd-d720-4321-a99f-ff31d7c0d27f [05:10] [ComfyUI] Calling execution.validate_prompt()... [05:10] [ComfyUI] Validation result: valid=True [05:10] [ComfyUI] === Validation PASSED === [05:10] [ComfyUI] got prompt [05:10] [ComfyUI] [comfy-env] Running DownloadAndLoadDepthAnythingV3Model: model: 'da3_large.safetensors', precision: 'auto', attention: 'auto' [05:10] [ComfyUI] [VRAM] Before DownloadAndLoadDepthAnythingV3Model: 302 / 24124 MB [05:10] [ComfyUI] [SubprocessWorker] call_method: nodes.load_model.DownloadAndLoadDepthAnythingV3Model.EXECUTE_NORMALIZED [05:10] [ComfyUI] [SubprocessWorker] acquired lock, ensuring started... [05:13] Frame 2 saved (frame_001.jpg, t=2.5s) [05:13] [capture-periodic] freeze=3ms shot=200ms unfreeze=5ms saved=True [05:13] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] WARNING: WARNING: You need pytorch with cu130 or higher to use optimized CUDA operations. [05:13] [ComfyUI] [SubprocessWorker] worker started/confirmed [05:13] [ComfyUI] [SubprocessWorker] serializing kwargs to shm... [05:13] [ComfyUI] [SubprocessWorker] created 0 shm blocks [05:13] [ComfyUI] [SubprocessWorker] sending request via socket... [05:13] [ComfyUI] [SubprocessWorker] call_id=1 sending call_method [05:14] [ComfyUI] Warning: You are sending unauthenticated requests to the HF Hub. Please set a HF_TOKEN to enable higher rate limits and faster downloads. [05:14] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] WARNING: Warning: You are sending unauthenticated requests to the HF Hub. Please set a HF_TOKEN to enable higher rate limits and faster downloads. [05:15] [ComfyUI] [05:15] Frame 3 saved (frame_002.jpg, t=4.8s) [05:15] [capture-periodic] freeze=3ms shot=153ms unfreeze=3ms saved=True [05:16] [capture-loop] iter=50 t=6.2s state={'complete': False, 'error': None, 'executedCount': 0, 'wsState': 1} eval_ms=4 ram=3.9/31.2GB (12%) swap=0.5/32.0GB (2%) cpu=3% browser_rss=0.85GB [05:18] [ComfyUI] model.safetensors: 0%| | 0.00/1.64G [00:00 cpu: 3623 / 24124 MB [05:40] Capturing execution frames: advanced_3d.json [05:40] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ping (timeout=5.0s)... [05:40] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ok [05:40] [ComfyUI] [SubprocessWorker] call_id=3 sending model_to_device [05:40] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] -290MB (now 591MB) reserved=2976MB peak=1172MB [05:41] [ComfyUI] [comfy-env] [VRAM] After offload 'DepthAnything3Net_1' (783 MB): 711 / 24124 MB [05:47] Frame 1 saved (frame_000.jpg, t=0.0s) [05:47] Validating workflow... [05:47] [ComfyUI] === /validate endpoint called === [05:47] [ComfyUI] Received JSON with keys: ['prompt'] [05:47] Queuing workflow for execution... [05:47] [ComfyUI] Prompt has 6 nodes [05:47] [ComfyUI] Node 2: LoadImage [05:47] [ComfyUI] Inputs: {"image": "kaffi.jpg"} [05:47] [ComfyUI] Node 3: DownloadAndLoadDepthAnythingV3Model [05:47] [ComfyUI] Inputs: {"model": "da3_large.safetensors", "precision": "auto", "attention": "auto"} [05:47] [ComfyUI] Node 21: DepthAnything_V3 [05:47] [ComfyUI] Inputs: {"normalization_mode": "Raw", "resize_method": "resize", "invert_depth": false, "keep_model_size": false, "da3_model": ["3", 0], "images": ["2", 0]} [05:47] [ComfyUI] Node 22: DA3_ToPointCloud [05:47] [ComfyUI] Inputs: {"confidence_threshold": 0.2, "downsample": 1, "allow_around_1": 1, "filter_outliers": false, "outlier_percentage": 5, "depth_raw": ["21", 0], "confidence": ["21", 1], "intrinsics": ["21", 6], "sky_ma [05:47] [ComfyUI] Node 23: DA3_SavePointCloud [05:47] [ComfyUI] Inputs: {"filename_prefix": "pointcloud", "pointcloud": ["22", 0]} [05:47] [ComfyUI] Node 25: DA3_PreviewPointCloud [05:47] [ComfyUI] Inputs: {"file_path": ["23", 0], "color_mode": "RGB", "preview_vtk": ""} [05:47] [ComfyUI] Generated prompt_id: e812d6fe-b4f3-4d6e-8699-c4d41e411f53 [05:47] [ComfyUI] Calling execution.validate_prompt()... [05:47] [ComfyUI] Validation result: valid=True [05:47] [ComfyUI] === Validation PASSED === [05:47] [ComfyUI] got prompt [05:47] [ComfyUI] [comfy-env] Running DownloadAndLoadDepthAnythingV3Model: model: 'da3_large.safetensors', precision: 'auto', attention: 'auto' [05:47] [ComfyUI] [VRAM] Before DownloadAndLoadDepthAnythingV3Model: 711 / 24124 MB [05:47] [ComfyUI] [ComfyUI-DepthAnythingV3] python: /workspaces/DepthAnythingV3-0759/ComfyUI/.ce/.pixi/envs/depthanythingv3-nodes/bin/python [05:47] [ComfyUI] [SubprocessWorker] call_method: nodes.load_model.DownloadAndLoadDepthAnythingV3Model.EXECUTE_NORMALIZED [05:47] [ComfyUI] [ComfyUI-DepthAnythingV3] sys_path sent to worker: ['/workspaces/DepthAnythingV3-0759/ComfyUI', '/workspaces/DepthAnythingV3-0759/ComfyUI/custom_nodes/ComfyUI-DepthAnythingV3', '/workspaces/DepthAnythingV3-0759/ComfyUI/.ce/.pixi/envs/depthanythingv3-nodes/lib/python3.1/site-packages', '/workspaces/DepthAnythingV3-0759/ComfyUI/custom_nodes/ComfyUI-DepthAnythingV3'] [05:47] [ComfyUI] [SubprocessWorker] acquired lock, ensuring started... [05:47] [ComfyUI] [SubprocessWorker] is_pixi=True, python=/workspaces/DepthAnythingV3-0759/ComfyUI/.ce/.pixi/envs/depthanythingv3-nodes/bin/python [05:47] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ping (timeout=5.0s)... [05:47] [ComfyUI] [SubprocessWorker] launching cmd=['/root/.pixi/bin/pixi', 'run', '--as-is', '--manifest-path', '/workspaces/DepthAnythingV3-0759/ComfyUI/.ce/pixi.toml', '-e', 'depthanythingv3-nodes', 'python', '/tmp/comfyui_pvenv_h_nkff2u/persistent_worker.py', 'abstract://comfy_worker_edfdb6ef60c5']... [05:47] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: failed [05:47] [ComfyUI] [SubprocessWorker] socket_addr=abstract://comfy_worker_edfdb6ef60c5 (abstract, no filesystem path) [05:47] [ComfyUI] [ComfyUI-DepthAnythingV3] Socket unhealthy, restarting worker... [05:47] [ComfyUI] [comfy-env] Invalidated 1 stale model patchers (will be cleaned up during next unload) [05:50] Frame 2 saved (frame_001.jpg, t=2.5s) [05:50] [capture-periodic] freeze=3ms shot=200ms unfreeze=6ms saved=True [05:50] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] WARNING: WARNING: You need pytorch with cu130 or higher to use optimized CUDA operations. [05:50] [ComfyUI] [SubprocessWorker] worker started/confirmed [05:50] [ComfyUI] [SubprocessWorker] serializing kwargs to shm... [05:50] [ComfyUI] [SubprocessWorker] created 0 shm blocks [05:50] [ComfyUI] [SubprocessWorker] sending request via socket... [05:50] [ComfyUI] [SubprocessWorker] call_id=4 sending call_method [05:51] [ComfyUI] [SubprocessWorker] got response: ok [05:51] [ComfyUI] [comfy-env] [VRAM] Before offload 'DepthAnything3Net_1' (783 MB) -> cpu: 563 / 24124 MB [05:51] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ping (timeout=5.0s)... [05:51] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ok [05:51] [ComfyUI] [SubprocessWorker] call_id=5 sending model_to_device [05:51] [ComfyUI] Model 'DepthAnything3Net_1' gone from worker (restarted?), treating as offloaded [05:51] [ComfyUI] [comfy-env] [VRAM] After offload 'DepthAnything3Net_1' (783 MB): 563 / 24124 MB [05:51] [ComfyUI] [comfy-env] DownloadAndLoadDepthAnythingV3Model done (3.45s): result: NodeOutput [05:51] [ComfyUI] [VRAM] After DownloadAndLoadDepthAnythingV3Model: 563 / 24124 MB [05:51] [ComfyUI] [comfy-env] Running DepthAnything_V3: normalization_mode: 'Raw', resize_method: 'resize', invert_depth: False, keep_model_size: False, da3_model: dict, images: torch.float32 [1,533,800,3] cpu [05:51] [ComfyUI] [VRAM] Before DepthAnything_V3: 563 / 24124 MB [05:51] [ComfyUI] [SubprocessWorker] call_method: nodes.nodes_inference.DepthAnything_V3.EXECUTE_NORMALIZED [05:51] [ComfyUI] [SubprocessWorker] acquired lock, ensuring started... [05:51] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ping (timeout=5.0s)... [05:51] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ok [05:51] [ComfyUI] [SubprocessWorker] worker started/confirmed [05:51] [ComfyUI] [comfy-env] PRE-SERIALIZE 'images' shape: torch.Size([1, 533, 800, 3]) [05:51] [ComfyUI] [SubprocessWorker] serializing kwargs to shm... [05:51] [ComfyUI] [SubprocessWorker] created 0 shm blocks [05:51] [ComfyUI] [SubprocessWorker] sending request via socket... [05:51] [ComfyUI] [SubprocessWorker] call_id=6 sending call_method [05:51] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] _from_shm got dict with keys: ['__type__', 'strategy', 'parent_pid', 'fd', 'storage_size'] [05:51] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZE TensorRef: tensor_size=[1, 533, 800, 3] [05:51] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZED tensor shape: torch.Size([1, 533, 800, 3]) [05:52] Frame 3 saved (frame_002.jpg, t=4.8s) [05:52] [capture-periodic] freeze=3ms shot=195ms unfreeze=7ms saved=True [05:52] [ComfyUI] [SubprocessWorker] callback 'request_vram_budget' from call_id=6 [05:52] [ComfyUI] [comfy-env] VRAM request: 0.82GB, free before eviction: 24.71GB [05:52] [ComfyUI] [comfy-env] VRAM after eviction: 24.71GB [05:52] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] +474MB (now 474MB) reserved=480MB peak=474MB [05:53] [capture-loop] iter=50 t=6.2s state={'complete': False, 'error': None, 'executedCount': 0, 'wsState': 1} eval_ms=3 ram=4.2/31.2GB (14%) swap=0.5/32.0GB (2%) cpu=4% browser_rss=0.94GB [05:53] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] +355MB (now 829MB) reserved=864MB peak=829MB [05:54] [ComfyUI] [SubprocessWorker] callback 'report_progress' from call_id=6 [05:54] [ComfyUI] [SubprocessWorker] got response: ok [05:54] [ComfyUI] [comfy-env] Created 1 model patchers: ['DepthAnything3Net_1'] [05:54] [ComfyUI] [comfy-env] DepthAnything_V3 done (3.66s): result: NodeOutput [05:54] [ComfyUI] [VRAM] After DepthAnything_V3: 2535 / 24124 MB [05:54] [ComfyUI] [VRAM] Loaded: SubprocessModel (783 MB) [05:54] [ComfyUI] [comfy-env] Running DA3_ToPointCloud: confidence_threshold: 0.2, downsample: 1, allow_around_1: True, filter_outliers: False, outlier_percentage: 5.0, depth_raw: torch.float32 [1,532,800,3] cpu, confidence: torch.float32 [1,532,800,3] cpu, intrinsics: '{\n "intrinsics": [\n {\n "image_0": [\n ..., sky_mask: torch.float32 [1,532,800] cpu, source_image: torch.float32 [1,532,800,3] cpu [05:54] [ComfyUI] [VRAM] Before DA3_ToPointCloud: 2535 / 24124 MB [05:54] [ComfyUI] [VRAM] Loaded: SubprocessModel (783 MB) [05:54] [ComfyUI] [SubprocessWorker] call_method: nodes.nodes_3d.DA3_ToPointCloud.EXECUTE_NORMALIZED [05:54] [ComfyUI] [SubprocessWorker] acquired lock, ensuring started... [05:54] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ping (timeout=5.0s)... [05:54] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ok [05:54] [ComfyUI] [SubprocessWorker] worker started/confirmed [05:54] [ComfyUI] [comfy-env] PRE-SERIALIZE 'depth_raw' shape: torch.Size([1, 532, 800, 3]) [05:54] [ComfyUI] [comfy-env] PRE-SERIALIZE 'confidence' shape: torch.Size([1, 532, 800, 3]) [05:54] [ComfyUI] [comfy-env] PRE-SERIALIZE 'sky_mask' shape: torch.Size([1, 532, 800]) [05:54] [ComfyUI] [comfy-env] PRE-SERIALIZE 'source_image' shape: torch.Size([1, 532, 800, 3]) [05:54] [ComfyUI] [SubprocessWorker] serializing kwargs to shm... [05:54] [ComfyUI] [SubprocessWorker] created 0 shm blocks [05:54] [ComfyUI] [SubprocessWorker] sending request via socket... [05:54] [ComfyUI] [SubprocessWorker] call_id=7 sending call_method [05:54] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] _from_shm got dict with keys: ['__type__', 'strategy', 'parent_pid', 'fd', 'storage_size'] [05:54] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZE TensorRef: tensor_size=[1, 532, 800, 3] [05:54] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZED tensor shape: torch.Size([1, 532, 800, 3]) [05:54] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] _from_shm got dict with keys: ['__type__', 'strategy', 'parent_pid', 'fd', 'storage_size'] [05:54] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZE TensorRef: tensor_size=[1, 532, 800, 3] [05:54] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZED tensor shape: torch.Size([1, 532, 800, 3]) [05:55] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] _from_shm got dict with keys: ['__type__', 'strategy', 'parent_pid', 'fd', 'storage_size'] [05:55] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZE TensorRef: tensor_size=[1, 532, 800] [05:55] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZED tensor shape: torch.Size([1, 532, 800]) [05:55] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] _from_shm got dict with keys: ['__type__', 'strategy', 'parent_pid', 'fd', 'storage_size'] [05:55] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZE TensorRef: tensor_size=[1, 532, 800, 3] [05:55] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZED tensor shape: torch.Size([1, 532, 800, 3]) [05:55] [ComfyUI] [SubprocessWorker] got response: ok [05:55] [ComfyUI] [comfy-env] DA3_ToPointCloud done (0.25s): result: NodeOutput [05:55] [ComfyUI] [VRAM] After DA3_ToPointCloud: 2535 / 24124 MB [05:55] [ComfyUI] [VRAM] Loaded: SubprocessModel (783 MB) [05:55] [ComfyUI] [comfy-env] Running DA3_SavePointCloud: filename_prefix: 'pointcloud', pointcloud: list[1] [05:55] [ComfyUI] [VRAM] Before DA3_SavePointCloud: 2535 / 24124 MB [05:55] [ComfyUI] [VRAM] Loaded: SubprocessModel (783 MB) [05:55] [ComfyUI] [SubprocessWorker] call_method: nodes.nodes_3d.DA3_SavePointCloud.EXECUTE_NORMALIZED [05:55] [ComfyUI] [SubprocessWorker] acquired lock, ensuring started... [05:55] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ping (timeout=5.0s)... [05:55] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ok [05:55] [ComfyUI] [SubprocessWorker] worker started/confirmed [05:55] [ComfyUI] [SubprocessWorker] serializing kwargs to shm... [05:55] [ComfyUI] [SubprocessWorker] created 0 shm blocks [05:55] [ComfyUI] [SubprocessWorker] sending request via socket... [05:55] [ComfyUI] [SubprocessWorker] call_id=8 sending call_method [05:55] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] _from_shm got dict with keys: ['__type__', 'strategy', 'parent_pid', 'fd', 'storage_size'] [05:55] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZE TensorRef: tensor_size=[310685, 3] [05:55] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZED tensor shape: torch.Size([310685, 3]) [05:55] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] _from_shm got dict with keys: ['__type__', 'strategy', 'parent_pid', 'fd', 'storage_size'] [05:55] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZE TensorRef: tensor_size=[310685] [05:55] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZED tensor shape: torch.Size([310685]) [05:55] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] _from_shm got dict with keys: ['__type__', 'strategy', 'parent_pid', 'fd', 'storage_size'] [05:55] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZE TensorRef: tensor_size=[310685, 3] [05:55] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZED tensor shape: torch.Size([310685, 3]) [05:56] Frame 4 saved (frame_003.jpg, t=9.3s) [05:56] [capture-periodic] freeze=3ms shot=197ms unfreeze=6ms saved=True [05:58] [ComfyUI] [SubprocessWorker] got response: ok [05:58] [ComfyUI] [comfy-env] DA3_SavePointCloud done (3.26s): result: NodeOutput [05:58] [ComfyUI] [VRAM] After DA3_SavePointCloud: 2535 / 24124 MB [05:58] [ComfyUI] [VRAM] Loaded: SubprocessModel (783 MB) [05:58] [ComfyUI] [comfy-env] Running DA3_PreviewPointCloud: file_path: '/workspaces/DepthAnythingV3-0759/ComfyUI/output/pointclo..., color_mode: 'RGB' [05:58] [ComfyUI] [VRAM] Before DA3_PreviewPointCloud: 2535 / 24124 MB [05:58] [ComfyUI] [VRAM] Loaded: SubprocessModel (783 MB) [05:58] [ComfyUI] [SubprocessWorker] call_method: nodes.preview_nodes.DA3_PreviewPointCloud.EXECUTE_NORMALIZED [05:58] [ComfyUI] [SubprocessWorker] acquired lock, ensuring started... [05:58] Node executed (1 total), capturing... [05:58] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ping (timeout=5.0s)... [05:58] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ok [05:58] [ComfyUI] [SubprocessWorker] worker started/confirmed [05:58] [ComfyUI] [SubprocessWorker] serializing kwargs to shm... [05:58] [ComfyUI] [SubprocessWorker] created 0 shm blocks [05:58] [ComfyUI] [SubprocessWorker] sending request via socket... [05:58] [ComfyUI] [SubprocessWorker] call_id=9 sending call_method [05:58] [ComfyUI] [SubprocessWorker] got response: ok [05:58] [ComfyUI] [comfy-env] DA3_PreviewPointCloud done (0.06s): result: NodeOutput [05:58] [ComfyUI] [VRAM] After DA3_PreviewPointCloud: 2535 / 24124 MB [05:58] [ComfyUI] [VRAM] Loaded: SubprocessModel (783 MB) [05:58] [ComfyUI] Prompt executed in 10.74 seconds [05:58] [ComfyUI] [comfy-env] [VRAM] Before offload 'DepthAnything3Net_1' (783 MB) -> cpu: 2535 / 24124 MB [05:58] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ping (timeout=5.0s)... [05:58] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ok [05:58] [ComfyUI] [SubprocessWorker] call_id=10 sending model_to_device [05:58] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] -366MB (now 463MB) reserved=1888MB peak=1219MB [05:59] [ComfyUI] [comfy-env] [VRAM] After offload 'DepthAnything3Net_1' (783 MB): 711 / 24124 MB [06:00] Frame 5 saved (frame_004.jpg, t=11.2s) [06:00] [capture-node] freeze=37ms shot=1647ms unfreeze=3ms saved=True [06:00] Node executed (2 total), capturing... [06:00] [capture-node] freeze=3ms shot=184ms unfreeze=0ms saved=False [06:00] Execution complete (t=13.2s) [06:00] Captured 5 unique frames over 13.41s [06:00] [debug] iframes: 1 [06:00] [debug] iframe[0]: src=http://127.0.0.1:41889/extensions/ComfyUI-GeometryPack/viewer_vtk.html?v=1778400, canvas=486x394, rafFrozen=True [06:00] [timing] trigger_3d_previews: 0.0s [06:03] Saved high-quality screenshot: advanced_3d_executed.png [06:03] [timing] final_screenshot: 0.5s [06:03] Capturing execution frames: advanced_3d_multiview.json [06:10] Frame 1 saved (frame_000.jpg, t=0.0s) [06:10] Validating workflow... [06:10] [ComfyUI] === /validate endpoint called === [06:10] [ComfyUI] Received JSON with keys: ['prompt'] [06:10] [ComfyUI] Prompt has 8 nodes [06:10] Queuing workflow for execution... [06:10] [ComfyUI] Node 3: DownloadAndLoadDepthAnythingV3Model [06:10] [ComfyUI] Inputs: {"model": "da3_large.safetensors", "precision": "auto", "attention": "auto"} [06:10] [ComfyUI] Node 21: LoadImageDataSetFromFolder [06:10] [ComfyUI] Inputs: {"folder": "SOH"} [06:10] [ComfyUI] Node 27: PreviewImage [06:10] [ComfyUI] Inputs: {"images": ["21", 0]} [06:10] [ComfyUI] Node 38: DA3_MultiViewPointCloud [06:10] [ComfyUI] Inputs: {"extrinsics": ["39", 4], "intrinsics": ["39", 5], "confidence_threshold": 0.3, "downsample": 4, "use_icp": false, "allow_around_1": false, "filter_outliers": true, "outlier_percentage": 5, "depths": [06:10] [ComfyUI] Node 39: DepthAnythingV3_MultiView [06:10] [ComfyUI] Inputs: {"normalization_mode": "Raw", "resize_method": "resize", "invert_depth": false, "keep_model_size": false, "da3_model": ["3", 0], "images": ["21", 0]} [06:10] [ComfyUI] Node 44: DA3_SavePointCloud [06:10] [ComfyUI] Inputs: {"filename_prefix": "pointcloud", "pointcloud": ["38", 0]} [06:10] [ComfyUI] Node 45: PreviewAny [06:10] [ComfyUI] Inputs: {"source": ["44", 0]} [06:10] [ComfyUI] Node 48: DA3_PreviewPointCloud [06:10] [ComfyUI] Inputs: {"file_path": ["44", 0], "color_mode": "RGB", "preview_vtk": ""} [06:10] [ComfyUI] Generated prompt_id: ae126a2d-8461-4704-8edd-85a11d11c310 [06:10] [ComfyUI] Calling execution.validate_prompt()... [06:10] [ComfyUI] Validation result: valid=True [06:10] [ComfyUI] === Validation PASSED === [06:10] [ComfyUI] got prompt [06:11] [ComfyUI] [comfy-env] Running DownloadAndLoadDepthAnythingV3Model: model: 'da3_large.safetensors', precision: 'auto', attention: 'auto' [06:11] [ComfyUI] [VRAM] Before DownloadAndLoadDepthAnythingV3Model: 711 / 24124 MB [06:11] [ComfyUI] [ComfyUI-DepthAnythingV3] python: /workspaces/DepthAnythingV3-0759/ComfyUI/.ce/.pixi/envs/depthanythingv3-nodes/bin/python [06:11] [ComfyUI] [SubprocessWorker] call_method: nodes.load_model.DownloadAndLoadDepthAnythingV3Model.EXECUTE_NORMALIZED [06:11] [ComfyUI] [ComfyUI-DepthAnythingV3] sys_path sent to worker: ['/workspaces/DepthAnythingV3-0759/ComfyUI', '/workspaces/DepthAnythingV3-0759/ComfyUI/custom_nodes/ComfyUI-DepthAnythingV3', '/workspaces/DepthAnythingV3-0759/ComfyUI/.ce/.pixi/envs/depthanythingv3-nodes/lib/python3.1/site-packages', '/workspaces/DepthAnythingV3-0759/ComfyUI/custom_nodes/ComfyUI-DepthAnythingV3'] [06:11] [ComfyUI] [SubprocessWorker] acquired lock, ensuring started... [06:11] [ComfyUI] [SubprocessWorker] is_pixi=True, python=/workspaces/DepthAnythingV3-0759/ComfyUI/.ce/.pixi/envs/depthanythingv3-nodes/bin/python [06:11] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ping (timeout=5.0s)... [06:11] [ComfyUI] [SubprocessWorker] launching cmd=['/root/.pixi/bin/pixi', 'run', '--as-is', '--manifest-path', '/workspaces/DepthAnythingV3-0759/ComfyUI/.ce/pixi.toml', '-e', 'depthanythingv3-nodes', 'python', '/tmp/comfyui_pvenv_h_nkff2u/persistent_worker.py', 'abstract://comfy_worker_7d6d0ef5234b']... [06:11] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: failed [06:11] [ComfyUI] [SubprocessWorker] socket_addr=abstract://comfy_worker_7d6d0ef5234b (abstract, no filesystem path) [06:11] [ComfyUI] [ComfyUI-DepthAnythingV3] Socket unhealthy, restarting worker... [06:11] [ComfyUI] [comfy-env] Invalidated 1 stale model patchers (will be cleaned up during next unload) [06:11] Node executed (1 total), capturing... [06:11] Frame 2 saved (frame_001.jpg, t=0.8s) [06:11] [capture-node] freeze=4ms shot=197ms unfreeze=7ms saved=True [06:14] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] WARNING: WARNING: You need pytorch with cu130 or higher to use optimized CUDA operations. [06:14] [ComfyUI] [SubprocessWorker] worker started/confirmed [06:14] [ComfyUI] [SubprocessWorker] serializing kwargs to shm... [06:14] [ComfyUI] [SubprocessWorker] created 0 shm blocks [06:14] [ComfyUI] [SubprocessWorker] sending request via socket... [06:14] [ComfyUI] [SubprocessWorker] call_id=11 sending call_method [06:14] [ComfyUI] [SubprocessWorker] got response: ok [06:14] [ComfyUI] [comfy-env] [VRAM] Before offload 'DepthAnything3Net_1' (783 MB) -> cpu: 563 / 24124 MB [06:14] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ping (timeout=5.0s)... [06:14] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ok [06:14] [ComfyUI] [SubprocessWorker] call_id=12 sending model_to_device [06:14] [ComfyUI] Model 'DepthAnything3Net_1' gone from worker (restarted?), treating as offloaded [06:14] [ComfyUI] [comfy-env] [VRAM] After offload 'DepthAnything3Net_1' (783 MB): 563 / 24124 MB [06:14] [ComfyUI] [comfy-env] DownloadAndLoadDepthAnythingV3Model done (3.60s): result: NodeOutput [06:14] [ComfyUI] [VRAM] After DownloadAndLoadDepthAnythingV3Model: 563 / 24124 MB [06:14] [ComfyUI] [comfy-env] Running DepthAnythingV3_MultiView: normalization_mode: 'Raw', resize_method: 'resize', invert_depth: False, keep_model_size: False, da3_model: dict, images: torch.float32 [1,680,1208,3] cpu [06:14] [ComfyUI] [VRAM] Before DepthAnythingV3_MultiView: 563 / 24124 MB [06:14] [ComfyUI] [SubprocessWorker] call_method: nodes.nodes_multiview.DepthAnythingV3_MultiView.EXECUTE_NORMALIZED [06:14] [ComfyUI] [SubprocessWorker] acquired lock, ensuring started... [06:14] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ping (timeout=5.0s)... [06:14] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ok [06:14] [ComfyUI] [SubprocessWorker] worker started/confirmed [06:14] [ComfyUI] [comfy-env] PRE-SERIALIZE 'images' shape: torch.Size([1, 680, 1208, 3]) [06:14] [ComfyUI] [SubprocessWorker] serializing kwargs to shm... [06:14] [ComfyUI] [SubprocessWorker] created 0 shm blocks [06:14] [ComfyUI] [SubprocessWorker] sending request via socket... [06:14] [ComfyUI] [SubprocessWorker] call_id=13 sending call_method [06:14] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] _from_shm got dict with keys: ['__type__', 'strategy', 'parent_pid', 'fd', 'storage_size'] [06:14] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZE TensorRef: tensor_size=[1, 680, 1208, 3] [06:14] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZED tensor shape: torch.Size([1, 680, 1208, 3]) [06:16] Frame 3 saved (frame_002.jpg, t=5.3s) [06:16] [capture-periodic] freeze=3ms shot=197ms unfreeze=7ms saved=True [06:16] [ComfyUI] [SubprocessWorker] callback 'request_vram_budget' from call_id=13 [06:16] [ComfyUI] [comfy-env] VRAM request: 0.82GB, free before eviction: 24.71GB [06:16] [ComfyUI] [comfy-env] VRAM after eviction: 24.71GB [06:16] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] +632MB (now 632MB) reserved=640MB peak=632MB [06:17] [capture-loop] iter=50 t=6.6s state={'complete': False, 'error': None, 'executedCount': 1, 'wsState': 1} eval_ms=4 ram=4.4/31.2GB (14%) swap=0.5/32.0GB (2%) cpu=6% browser_rss=1.00GB [06:17] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] +231MB (now 863MB) reserved=896MB peak=863MB [06:18] [ComfyUI] [SubprocessWorker] got response: ok [06:18] [ComfyUI] [comfy-env] Created 1 model patchers: ['DepthAnything3Net_1'] [06:18] [ComfyUI] [comfy-env] DepthAnythingV3_MultiView done (3.90s): result: NodeOutput [06:18] [ComfyUI] [VRAM] After DepthAnythingV3_MultiView: 4007 / 24124 MB [06:18] [ComfyUI] [VRAM] Loaded: SubprocessModel (783 MB) [06:18] [ComfyUI] [comfy-env] Running DepthAnythingV3_MultiView: normalization_mode: 'Raw', resize_method: 'resize', invert_depth: False, keep_model_size: False, da3_model: dict, images: torch.float32 [1,680,1208,3] cpu [06:18] [ComfyUI] [VRAM] Before DepthAnythingV3_MultiView: 4007 / 24124 MB [06:18] [ComfyUI] [VRAM] Loaded: SubprocessModel (783 MB) [06:18] [ComfyUI] [SubprocessWorker] call_method: nodes.nodes_multiview.DepthAnythingV3_MultiView.EXECUTE_NORMALIZED [06:18] [ComfyUI] [SubprocessWorker] acquired lock, ensuring started... [06:18] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ping (timeout=5.0s)... [06:18] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ok [06:18] [ComfyUI] [SubprocessWorker] worker started/confirmed [06:18] [ComfyUI] [comfy-env] PRE-SERIALIZE 'images' shape: torch.Size([1, 680, 1208, 3]) [06:18] [ComfyUI] [SubprocessWorker] serializing kwargs to shm... [06:18] [ComfyUI] [SubprocessWorker] created 0 shm blocks [06:18] [ComfyUI] [SubprocessWorker] sending request via socket... [06:18] [ComfyUI] [SubprocessWorker] call_id=14 sending call_method [06:18] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] _from_shm got dict with keys: ['__type__', 'strategy', 'parent_pid', 'fd', 'storage_size'] [06:18] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZE TensorRef: tensor_size=[1, 680, 1208, 3] [06:18] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZED tensor shape: torch.Size([1, 680, 1208, 3]) [06:18] [ComfyUI] [SubprocessWorker] callback 'request_vram_budget' from call_id=14 [06:18] [ComfyUI] [comfy-env] VRAM request: 0.82GB, free before eviction: 21.09GB [06:18] [ComfyUI] [comfy-env] VRAM after eviction: 21.09GB [06:18] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] +376MB (now 1240MB) reserved=1440MB peak=1662MB [06:19] [ComfyUI] [SubprocessWorker] got response: ok [06:19] [ComfyUI] [comfy-env] DepthAnythingV3_MultiView done (0.70s): result: NodeOutput [06:19] [ComfyUI] [VRAM] After DepthAnythingV3_MultiView: 3591 / 24124 MB [06:19] [ComfyUI] [VRAM] Loaded: SubprocessModel (783 MB) [06:19] [ComfyUI] [comfy-env] Running DA3_MultiViewPointCloud: extrinsics: '{\n "extrinsics": [\n {\n "image_0": [\n ..., intrinsics: '{\n "intrinsics": [\n {\n "image_0": [\n ..., confidence_threshold: 0.3, downsample: 4, use_icp: False, allow_around_1: False, filter_outliers: True, outlier_percentage: 5.0, depths: torch.float32 [1,680,1208,3] cpu, images: torch.float32 [1,680,1208,3] cpu, confidence: torch.float32 [1,680,1208,3] cpu, sky_mask: torch.float32 [1,680,1208] cpu [06:19] [ComfyUI] [VRAM] Before DA3_MultiViewPointCloud: 3591 / 24124 MB [06:19] [ComfyUI] [VRAM] Loaded: SubprocessModel (783 MB) [06:19] [ComfyUI] [SubprocessWorker] call_method: nodes.nodes_multiview.DA3_MultiViewPointCloud.EXECUTE_NORMALIZED [06:19] [ComfyUI] [SubprocessWorker] acquired lock, ensuring started... [06:19] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ping (timeout=5.0s)... [06:19] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ok [06:19] [ComfyUI] [SubprocessWorker] worker started/confirmed [06:19] [ComfyUI] [comfy-env] PRE-SERIALIZE 'depths' shape: torch.Size([1, 680, 1208, 3]) [06:19] [ComfyUI] [comfy-env] PRE-SERIALIZE 'images' shape: torch.Size([1, 680, 1208, 3]) [06:19] [ComfyUI] [comfy-env] PRE-SERIALIZE 'confidence' shape: torch.Size([1, 680, 1208, 3]) [06:19] [ComfyUI] [comfy-env] PRE-SERIALIZE 'sky_mask' shape: torch.Size([1, 680, 1208]) [06:19] [ComfyUI] [SubprocessWorker] serializing kwargs to shm... [06:19] [ComfyUI] [SubprocessWorker] created 0 shm blocks [06:19] [ComfyUI] [SubprocessWorker] sending request via socket... [06:19] [ComfyUI] [SubprocessWorker] call_id=15 sending call_method [06:19] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] _from_shm got dict with keys: ['__type__', 'strategy', 'parent_pid', 'fd', 'storage_size'] [06:19] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZE TensorRef: tensor_size=[1, 680, 1208, 3] [06:19] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZED tensor shape: torch.Size([1, 680, 1208, 3]) [06:19] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] _from_shm got dict with keys: ['__type__', 'strategy', 'parent_pid', 'fd', 'storage_size'] [06:19] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZE TensorRef: tensor_size=[1, 680, 1208, 3] [06:19] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZED tensor shape: torch.Size([1, 680, 1208, 3]) [06:19] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] _from_shm got dict with keys: ['__type__', 'strategy', 'parent_pid', 'fd', 'storage_size'] [06:19] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZE TensorRef: tensor_size=[1, 680, 1208, 3] [06:19] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZED tensor shape: torch.Size([1, 680, 1208, 3]) [06:19] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] _from_shm got dict with keys: ['__type__', 'strategy', 'parent_pid', 'fd', 'storage_size'] [06:19] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZE TensorRef: tensor_size=[1, 680, 1208] [06:19] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZED tensor shape: torch.Size([1, 680, 1208]) [06:19] [ComfyUI] [SubprocessWorker] callback 'report_progress' from call_id=15 [06:19] [ComfyUI] [SubprocessWorker] got response: ok [06:19] [ComfyUI] [comfy-env] DA3_MultiViewPointCloud done (0.25s): result: NodeOutput [06:19] [ComfyUI] [VRAM] After DA3_MultiViewPointCloud: 3591 / 24124 MB [06:19] [ComfyUI] [VRAM] Loaded: SubprocessModel (783 MB) [06:19] [ComfyUI] [comfy-env] Running DA3_MultiViewPointCloud: extrinsics: '{\n "extrinsics": [\n {\n "image_0": [\n ..., intrinsics: '{\n "intrinsics": [\n {\n "image_0": [\n ..., confidence_threshold: 0.3, downsample: 4, use_icp: False, allow_around_1: False, filter_outliers: True, outlier_percentage: 5.0, depths: torch.float32 [1,680,1208,3] cpu, images: torch.float32 [1,680,1208,3] cpu, confidence: torch.float32 [1,680,1208,3] cpu, sky_mask: torch.float32 [1,680,1208] cpu [06:19] [ComfyUI] [VRAM] Before DA3_MultiViewPointCloud: 3591 / 24124 MB [06:19] [ComfyUI] [VRAM] Loaded: SubprocessModel (783 MB) [06:19] [ComfyUI] [SubprocessWorker] call_method: nodes.nodes_multiview.DA3_MultiViewPointCloud.EXECUTE_NORMALIZED [06:19] [ComfyUI] [SubprocessWorker] acquired lock, ensuring started... [06:19] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ping (timeout=5.0s)... [06:19] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ok [06:19] [ComfyUI] [SubprocessWorker] worker started/confirmed [06:19] [ComfyUI] [comfy-env] PRE-SERIALIZE 'depths' shape: torch.Size([1, 680, 1208, 3]) [06:19] [ComfyUI] [comfy-env] PRE-SERIALIZE 'images' shape: torch.Size([1, 680, 1208, 3]) [06:19] [ComfyUI] [comfy-env] PRE-SERIALIZE 'confidence' shape: torch.Size([1, 680, 1208, 3]) [06:19] [ComfyUI] [comfy-env] PRE-SERIALIZE 'sky_mask' shape: torch.Size([1, 680, 1208]) [06:19] [ComfyUI] [SubprocessWorker] serializing kwargs to shm... [06:19] [ComfyUI] [SubprocessWorker] created 0 shm blocks [06:19] [ComfyUI] [SubprocessWorker] sending request via socket... [06:19] [ComfyUI] [SubprocessWorker] call_id=16 sending call_method [06:19] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] _from_shm got dict with keys: ['__type__', 'strategy', 'parent_pid', 'fd', 'storage_size'] [06:19] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZE TensorRef: tensor_size=[1, 680, 1208, 3] [06:19] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZED tensor shape: torch.Size([1, 680, 1208, 3]) [06:19] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] _from_shm got dict with keys: ['__type__', 'strategy', 'parent_pid', 'fd', 'storage_size'] [06:19] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZE TensorRef: tensor_size=[1, 680, 1208, 3] [06:19] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZED tensor shape: torch.Size([1, 680, 1208, 3]) [06:19] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] _from_shm got dict with keys: ['__type__', 'strategy', 'parent_pid', 'fd', 'storage_size'] [06:19] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZE TensorRef: tensor_size=[1, 680, 1208, 3] [06:19] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZED tensor shape: torch.Size([1, 680, 1208, 3]) [06:19] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] _from_shm got dict with keys: ['__type__', 'strategy', 'parent_pid', 'fd', 'storage_size'] [06:19] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZE TensorRef: tensor_size=[1, 680, 1208] [06:19] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZED tensor shape: torch.Size([1, 680, 1208]) [06:19] [ComfyUI] [SubprocessWorker] callback 'report_progress' from call_id=16 [06:19] [ComfyUI] [SubprocessWorker] got response: ok [06:19] [ComfyUI] [comfy-env] DA3_MultiViewPointCloud done (0.25s): result: NodeOutput [06:19] [ComfyUI] [VRAM] After DA3_MultiViewPointCloud: 3591 / 24124 MB [06:19] [ComfyUI] [VRAM] Loaded: SubprocessModel (783 MB) [06:19] [ComfyUI] [comfy-env] Running DA3_SavePointCloud: filename_prefix: 'pointcloud', pointcloud: list[1] [06:19] [ComfyUI] [VRAM] Before DA3_SavePointCloud: 3591 / 24124 MB [06:19] [ComfyUI] [VRAM] Loaded: SubprocessModel (783 MB) [06:19] [ComfyUI] [SubprocessWorker] call_method: nodes.nodes_3d.DA3_SavePointCloud.EXECUTE_NORMALIZED [06:19] [ComfyUI] [SubprocessWorker] acquired lock, ensuring started... [06:19] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ping (timeout=5.0s)... [06:19] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ok [06:19] [ComfyUI] [SubprocessWorker] worker started/confirmed [06:19] [ComfyUI] [SubprocessWorker] serializing kwargs to shm... [06:19] [ComfyUI] [SubprocessWorker] created 0 shm blocks [06:19] [ComfyUI] [SubprocessWorker] sending request via socket... [06:19] [ComfyUI] [SubprocessWorker] call_id=17 sending call_method [06:19] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] _from_shm got dict with keys: ['__type__', 'strategy', 'parent_pid', 'fd', 'storage_size'] [06:19] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZE TensorRef: tensor_size=[18824, 3] [06:19] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZED tensor shape: torch.Size([18824, 3]) [06:19] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] _from_shm got dict with keys: ['__type__', 'strategy', 'parent_pid', 'fd', 'storage_size'] [06:19] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZE TensorRef: tensor_size=[18824, 3] [06:19] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZED tensor shape: torch.Size([18824, 3]) [06:19] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] _from_shm got dict with keys: ['__type__', 'strategy', 'parent_pid', 'fd', 'storage_size'] [06:19] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZE TensorRef: tensor_size=[18824] [06:19] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZED tensor shape: torch.Size([18824]) [06:19] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] _from_shm got dict with keys: ['__type__', 'strategy', 'parent_pid', 'fd', 'storage_size'] [06:19] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZE TensorRef: tensor_size=[18824] [06:19] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZED tensor shape: torch.Size([18824]) [06:20] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] -448MB (now 793MB) reserved=2944MB peak=1662MB [06:20] [ComfyUI] [SubprocessWorker] got response: ok [06:20] [ComfyUI] [comfy-env] DA3_SavePointCloud done (0.50s): result: NodeOutput [06:20] [ComfyUI] [VRAM] After DA3_SavePointCloud: 3591 / 24124 MB [06:20] [ComfyUI] [VRAM] Loaded: SubprocessModel (783 MB) [06:20] [ComfyUI] [comfy-env] Running DA3_SavePointCloud: filename_prefix: 'pointcloud', pointcloud: list[1] [06:20] [ComfyUI] [VRAM] Before DA3_SavePointCloud: 3591 / 24124 MB [06:20] [ComfyUI] [VRAM] Loaded: SubprocessModel (783 MB) [06:20] [ComfyUI] [SubprocessWorker] call_method: nodes.nodes_3d.DA3_SavePointCloud.EXECUTE_NORMALIZED [06:20] [ComfyUI] [SubprocessWorker] acquired lock, ensuring started... [06:20] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ping (timeout=5.0s)... [06:20] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ok [06:20] [ComfyUI] [SubprocessWorker] worker started/confirmed [06:20] [ComfyUI] [SubprocessWorker] serializing kwargs to shm... [06:20] [ComfyUI] [SubprocessWorker] created 0 shm blocks [06:20] [ComfyUI] [SubprocessWorker] sending request via socket... [06:20] [ComfyUI] [SubprocessWorker] call_id=18 sending call_method [06:20] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] _from_shm got dict with keys: ['__type__', 'strategy', 'parent_pid', 'fd', 'storage_size'] [06:20] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZE TensorRef: tensor_size=[20457, 3] [06:20] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZED tensor shape: torch.Size([20457, 3]) [06:20] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] _from_shm got dict with keys: ['__type__', 'strategy', 'parent_pid', 'fd', 'storage_size'] [06:20] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZE TensorRef: tensor_size=[20457, 3] [06:20] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZED tensor shape: torch.Size([20457, 3]) [06:20] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] _from_shm got dict with keys: ['__type__', 'strategy', 'parent_pid', 'fd', 'storage_size'] [06:20] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZE TensorRef: tensor_size=[20457] [06:20] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZED tensor shape: torch.Size([20457]) [06:20] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] _from_shm got dict with keys: ['__type__', 'strategy', 'parent_pid', 'fd', 'storage_size'] [06:20] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZE TensorRef: tensor_size=[20457] [06:20] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZED tensor shape: torch.Size([20457]) [06:20] Frame 4 saved (frame_003.jpg, t=9.8s) [06:20] [capture-periodic] freeze=3ms shot=203ms unfreeze=7ms saved=True [06:20] [ComfyUI] [SubprocessWorker] got response: ok [06:20] [ComfyUI] [comfy-env] DA3_SavePointCloud done (0.50s): result: NodeOutput [06:20] [ComfyUI] [VRAM] After DA3_SavePointCloud: 3591 / 24124 MB [06:20] [ComfyUI] [VRAM] Loaded: SubprocessModel (783 MB) [06:20] [ComfyUI] [comfy-env] Running DA3_PreviewPointCloud: file_path: '/workspaces/DepthAnythingV3-0759/ComfyUI/output/pointclo..., color_mode: 'RGB' [06:20] [ComfyUI] [VRAM] Before DA3_PreviewPointCloud: 3591 / 24124 MB [06:20] [ComfyUI] [VRAM] Loaded: SubprocessModel (783 MB) [06:20] [ComfyUI] [SubprocessWorker] call_method: nodes.preview_nodes.DA3_PreviewPointCloud.EXECUTE_NORMALIZED [06:20] [ComfyUI] [SubprocessWorker] acquired lock, ensuring started... [06:20] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ping (timeout=5.0s)... [06:20] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ok [06:20] [ComfyUI] [SubprocessWorker] worker started/confirmed [06:20] [ComfyUI] [SubprocessWorker] serializing kwargs to shm... [06:20] [ComfyUI] [SubprocessWorker] created 0 shm blocks [06:20] [ComfyUI] [SubprocessWorker] sending request via socket... [06:20] Node executed (2 total), capturing... [06:20] [ComfyUI] [SubprocessWorker] call_id=19 sending call_method [06:20] [ComfyUI] [SubprocessWorker] got response: ok [06:20] [ComfyUI] [comfy-env] DA3_PreviewPointCloud done (0.07s): result: NodeOutput [06:20] [ComfyUI] [VRAM] After DA3_PreviewPointCloud: 3591 / 24124 MB [06:20] [ComfyUI] [VRAM] Loaded: SubprocessModel (783 MB) [06:20] [ComfyUI] [comfy-env] Running DA3_PreviewPointCloud: file_path: '/workspaces/DepthAnythingV3-0759/ComfyUI/output/pointclo..., color_mode: 'RGB' [06:20] [ComfyUI] [VRAM] Before DA3_PreviewPointCloud: 3591 / 24124 MB [06:20] [ComfyUI] [VRAM] Loaded: SubprocessModel (783 MB) [06:20] [ComfyUI] [SubprocessWorker] call_method: nodes.preview_nodes.DA3_PreviewPointCloud.EXECUTE_NORMALIZED [06:20] [ComfyUI] [SubprocessWorker] acquired lock, ensuring started... [06:20] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ping (timeout=5.0s)... [06:20] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ok [06:20] [ComfyUI] [SubprocessWorker] worker started/confirmed [06:20] [ComfyUI] [SubprocessWorker] serializing kwargs to shm... [06:20] [ComfyUI] [SubprocessWorker] created 0 shm blocks [06:20] [ComfyUI] [SubprocessWorker] sending request via socket... [06:20] [ComfyUI] [SubprocessWorker] call_id=20 sending call_method [06:20] [ComfyUI] [SubprocessWorker] got response: ok [06:20] [ComfyUI] [comfy-env] DA3_PreviewPointCloud done (0.07s): result: NodeOutput [06:20] [ComfyUI] [VRAM] After DA3_PreviewPointCloud: 3591 / 24124 MB [06:20] [ComfyUI] [VRAM] Loaded: SubprocessModel (783 MB) [06:20] [ComfyUI] Prompt executed in 10.11 seconds [06:21] Frame 5 saved (frame_004.jpg, t=10.5s) [06:21] [capture-node] freeze=7ms shot=209ms unfreeze=7ms saved=True [06:21] Node executed (4 total), capturing... [06:21] Frame 6 saved (frame_005.jpg, t=11.2s) [06:21] [capture-node] freeze=4ms shot=218ms unfreeze=0ms saved=True [06:21] Execution complete (t=11.2s) [06:21] Captured 6 unique frames over 11.56s [06:22] [debug] iframes: 1 [06:22] [debug] iframe[0]: src=http://127.0.0.1:41889/extensions/ComfyUI-DepthAnythingV3/viewer_vtk.html?v=1778, canvas=486x394, rafFrozen=True [06:22] [timing] trigger_3d_previews: 0.0s [06:24] Saved high-quality screenshot: advanced_3d_multiview_executed.png [06:24] [timing] final_screenshot: 0.5s [06:24] [ComfyUI] [comfy-env] [VRAM] Before offload 'DepthAnything3Net_1' (783 MB) -> cpu: 3591 / 24124 MB [06:24] Capturing execution frames: bas_relief.json [06:24] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ping (timeout=5.0s)... [06:24] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ok [06:24] [ComfyUI] [SubprocessWorker] call_id=21 sending model_to_device [06:24] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] -250MB (now 543MB) reserved=2944MB peak=1662MB [06:25] [ComfyUI] [comfy-env] [VRAM] After offload 'DepthAnything3Net_1' (783 MB): 711 / 24124 MB [06:32] Frame 1 saved (frame_000.jpg, t=0.0s) [06:32] Validating workflow... [06:32] [ComfyUI] === /validate endpoint called === [06:32] [ComfyUI] Received JSON with keys: ['prompt'] [06:32] [ComfyUI] Prompt has 6 nodes [06:32] Queuing workflow for execution... [06:32] [ComfyUI] Node 2: LoadImage [06:32] [ComfyUI] Inputs: {"image": "bas_relief.png"} [06:32] [ComfyUI] Node 3: DownloadAndLoadDepthAnythingV3Model [06:32] [ComfyUI] Inputs: {"model": "da3_large.safetensors", "precision": "auto", "attention": "auto"} [06:32] [ComfyUI] Node 21: DepthAnything_V3 [06:32] [ComfyUI] Inputs: {"normalization_mode": "Raw", "resize_method": "resize", "invert_depth": false, "keep_model_size": true, "da3_model": ["3", 0], "images": ["2", 0]} [06:32] [ComfyUI] Node 27: PreviewImage [06:32] [ComfyUI] Inputs: {"images": ["21", 2]} [06:32] [ComfyUI] Node 30: Preview3D [06:32] [ComfyUI] Inputs: {"model_file": ["32", 0], "image": ""} [06:32] [ComfyUI] Node 32: DA3_ToMesh [06:32] [ComfyUI] Inputs: {"confidence_threshold": 0.1, "depth_edge_threshold": 0.1, "downsample": 1, "target_faces": 100000, "filename_prefix": "mesh", "allow_around_1": true, "use_draco_compression": false, "depth_raw": ["21 [06:32] [ComfyUI] Generated prompt_id: fc1e5f23-25be-469e-bdc7-11af852c2b4e [06:32] [ComfyUI] Calling execution.validate_prompt()... [06:32] [ComfyUI] Validation result: valid=True [06:32] [ComfyUI] === Validation PASSED === [06:32] [ComfyUI] got prompt [06:32] [ComfyUI] [comfy-env] Running DownloadAndLoadDepthAnythingV3Model: model: 'da3_large.safetensors', precision: 'auto', attention: 'auto' [06:32] [ComfyUI] [VRAM] Before DownloadAndLoadDepthAnythingV3Model: 711 / 24124 MB [06:32] [ComfyUI] [ComfyUI-DepthAnythingV3] python: /workspaces/DepthAnythingV3-0759/ComfyUI/.ce/.pixi/envs/depthanythingv3-nodes/bin/python [06:32] [ComfyUI] [SubprocessWorker] call_method: nodes.load_model.DownloadAndLoadDepthAnythingV3Model.EXECUTE_NORMALIZED [06:32] [ComfyUI] [ComfyUI-DepthAnythingV3] sys_path sent to worker: ['/workspaces/DepthAnythingV3-0759/ComfyUI', '/workspaces/DepthAnythingV3-0759/ComfyUI/custom_nodes/ComfyUI-DepthAnythingV3', '/workspaces/DepthAnythingV3-0759/ComfyUI/.ce/.pixi/envs/depthanythingv3-nodes/lib/python3.1/site-packages', '/workspaces/DepthAnythingV3-0759/ComfyUI/custom_nodes/ComfyUI-DepthAnythingV3'] [06:32] [ComfyUI] [SubprocessWorker] acquired lock, ensuring started... [06:32] [ComfyUI] [SubprocessWorker] is_pixi=True, python=/workspaces/DepthAnythingV3-0759/ComfyUI/.ce/.pixi/envs/depthanythingv3-nodes/bin/python [06:32] [ComfyUI] [SubprocessWorker] launching cmd=['/root/.pixi/bin/pixi', 'run', '--as-is', '--manifest-path', '/workspaces/DepthAnythingV3-0759/ComfyUI/.ce/pixi.toml', '-e', 'depthanythingv3-nodes', 'python', '/tmp/comfyui_pvenv_h_nkff2u/persistent_worker.py', 'abstract://comfy_worker_a55411bac876']... [06:32] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ping (timeout=5.0s)... [06:32] [ComfyUI] [SubprocessWorker] socket_addr=abstract://comfy_worker_a55411bac876 (abstract, no filesystem path) [06:32] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: failed [06:32] [ComfyUI] [ComfyUI-DepthAnythingV3] Socket unhealthy, restarting worker... [06:32] [ComfyUI] [comfy-env] Invalidated 1 stale model patchers (will be cleaned up during next unload) [06:34] Frame 2 saved (frame_001.jpg, t=2.5s) [06:34] [capture-periodic] freeze=4ms shot=269ms unfreeze=3ms saved=True [06:35] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] WARNING: WARNING: You need pytorch with cu130 or higher to use optimized CUDA operations. [06:35] [ComfyUI] [SubprocessWorker] worker started/confirmed [06:35] [ComfyUI] [SubprocessWorker] serializing kwargs to shm... [06:35] [ComfyUI] [SubprocessWorker] created 0 shm blocks [06:35] [ComfyUI] [SubprocessWorker] sending request via socket... [06:35] [ComfyUI] [SubprocessWorker] call_id=22 sending call_method [06:35] [ComfyUI] [SubprocessWorker] got response: ok [06:35] [ComfyUI] [comfy-env] [VRAM] Before offload 'DepthAnything3Net_1' (783 MB) -> cpu: 563 / 24124 MB [06:35] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ping (timeout=5.0s)... [06:35] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ok [06:35] [ComfyUI] [SubprocessWorker] call_id=23 sending model_to_device [06:35] [ComfyUI] Model 'DepthAnything3Net_1' gone from worker (restarted?), treating as offloaded [06:35] [ComfyUI] [comfy-env] [VRAM] After offload 'DepthAnything3Net_1' (783 MB): 563 / 24124 MB [06:35] [ComfyUI] [comfy-env] DownloadAndLoadDepthAnythingV3Model done (3.45s): result: NodeOutput [06:35] [ComfyUI] [VRAM] After DownloadAndLoadDepthAnythingV3Model: 563 / 24124 MB [06:35] [ComfyUI] [comfy-env] Running DepthAnything_V3: normalization_mode: 'Raw', resize_method: 'resize', invert_depth: False, keep_model_size: True, da3_model: dict, images: torch.float32 [1,2000,1336,3] cpu [06:35] [ComfyUI] [VRAM] Before DepthAnything_V3: 563 / 24124 MB [06:35] [ComfyUI] [SubprocessWorker] call_method: nodes.nodes_inference.DepthAnything_V3.EXECUTE_NORMALIZED [06:35] [ComfyUI] [SubprocessWorker] acquired lock, ensuring started... [06:35] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ping (timeout=5.0s)... [06:35] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ok [06:35] [ComfyUI] [SubprocessWorker] worker started/confirmed [06:36] [ComfyUI] [comfy-env] PRE-SERIALIZE 'images' shape: torch.Size([1, 2000, 1336, 3]) [06:36] [ComfyUI] [SubprocessWorker] serializing kwargs to shm... [06:36] [ComfyUI] [SubprocessWorker] created 0 shm blocks [06:36] [ComfyUI] [SubprocessWorker] sending request via socket... [06:36] [ComfyUI] [SubprocessWorker] call_id=24 sending call_method [06:36] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] _from_shm got dict with keys: ['__type__', 'strategy', 'parent_pid', 'fd', 'storage_size'] [06:36] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZE TensorRef: tensor_size=[1, 2000, 1336, 3] [06:36] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZED tensor shape: torch.Size([1, 2000, 1336, 3]) [06:37] Frame 3 saved (frame_002.jpg, t=4.9s) [06:37] [capture-periodic] freeze=3ms shot=239ms unfreeze=7ms saved=True [06:37] [ComfyUI] [SubprocessWorker] callback 'request_vram_budget' from call_id=24 [06:37] [ComfyUI] [comfy-env] VRAM request: 0.82GB, free before eviction: 24.71GB [06:37] [ComfyUI] [comfy-env] VRAM after eviction: 24.71GB [06:37] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] +578MB (now 578MB) reserved=608MB peak=578MB [06:38] [capture-loop] iter=50 t=6.4s state={'complete': False, 'error': None, 'executedCount': 0, 'wsState': 1} eval_ms=3 ram=4.5/31.2GB (14%) swap=0.5/32.0GB (2%) cpu=5% browser_rss=0.91GB [06:38] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] +442MB (now 1021MB) reserved=1056MB peak=1021MB [06:39] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] +324MB (now 1345MB) reserved=1536MB peak=1345MB [06:40] [ComfyUI] [SubprocessWorker] callback 'report_progress' from call_id=24 [06:40] [ComfyUI] [SubprocessWorker] got response: ok [06:40] [ComfyUI] [comfy-env] Created 1 model patchers: ['DepthAnything3Net_1'] [06:40] [ComfyUI] [comfy-env] DepthAnything_V3 done (4.55s): result: NodeOutput [06:40] [ComfyUI] [VRAM] After DepthAnything_V3: 7655 / 24124 MB [06:40] [ComfyUI] [VRAM] Loaded: SubprocessModel (783 MB) [06:40] [ComfyUI] [comfy-env] Running DA3_ToMesh: confidence_threshold: 0.1, depth_edge_threshold: 0.1, downsample: 1, target_faces: 100000, filename_prefix: 'mesh', allow_around_1: True, use_draco_compression: False, depth_raw: torch.float32 [1,2002,1330,3] cpu, confidence: torch.float32 [1,2002,1330,3] cpu, intrinsics: '{\n "intrinsics": [\n {\n "image_0": [\n ..., sky_mask: torch.float32 [1,2002,1330] cpu, source_image: torch.float32 [1,2002,1330,3] cpu [06:40] [ComfyUI] [ComfyUI-DepthAnythingV3] python: /workspaces/DepthAnythingV3-0759/ComfyUI/.ce/.pixi/envs/depthanythingv3-nodes/bin/python [06:40] [ComfyUI] [VRAM] Before DA3_ToMesh: 7655 / 24124 MB [06:40] [ComfyUI] [VRAM] Loaded: SubprocessModel (783 MB) [06:40] [ComfyUI] [ComfyUI-DepthAnythingV3] sys_path sent to worker: ['/workspaces/DepthAnythingV3-0759/ComfyUI', '/workspaces/DepthAnythingV3-0759/ComfyUI/custom_nodes/ComfyUI-DepthAnythingV3', '/workspaces/DepthAnythingV3-0759/ComfyUI/.ce/.pixi/envs/depthanythingv3-nodes/lib/python3.1/site-packages', '/workspaces/DepthAnythingV3-0759/ComfyUI/custom_nodes/ComfyUI-DepthAnythingV3'] [06:40] [ComfyUI] [SubprocessWorker] call_method: nodes.nodes_3d.DA3_ToMesh.EXECUTE_NORMALIZED [06:40] [ComfyUI] [SubprocessWorker] is_pixi=True, python=/workspaces/DepthAnythingV3-0759/ComfyUI/.ce/.pixi/envs/depthanythingv3-nodes/bin/python [06:40] [ComfyUI] [SubprocessWorker] acquired lock, ensuring started... [06:40] [ComfyUI] [SubprocessWorker] launching cmd=['/root/.pixi/bin/pixi', 'run', '--as-is', '--manifest-path', '/workspaces/DepthAnythingV3-0759/ComfyUI/.ce/pixi.toml', '-e', 'depthanythingv3-nodes', 'python', '/tmp/comfyui_pvenv_h_nkff2u/persistent_worker.py', 'abstract://comfy_worker_580f6b668a66']... [06:40] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ping (timeout=5.0s)... [06:40] [ComfyUI] [SubprocessWorker] socket_addr=abstract://comfy_worker_580f6b668a66 (abstract, no filesystem path) [06:40] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: failed [06:40] [ComfyUI] [ComfyUI-DepthAnythingV3] Socket unhealthy, restarting worker... [06:40] [ComfyUI] [comfy-env] Invalidated 1 stale model patchers (will be cleaned up during next unload) [06:40] Node executed (1 total), capturing... [06:41] Frame 4 saved (frame_003.jpg, t=9.3s) [06:41] [capture-node] freeze=4ms shot=261ms unfreeze=4ms saved=True [06:44] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] WARNING: WARNING: You need pytorch with cu130 or higher to use optimized CUDA operations. [06:44] [ComfyUI] [SubprocessWorker] worker started/confirmed [06:44] [ComfyUI] [comfy-env] PRE-SERIALIZE 'depth_raw' shape: torch.Size([1, 2002, 1330, 3]) [06:44] [ComfyUI] [comfy-env] PRE-SERIALIZE 'confidence' shape: torch.Size([1, 2002, 1330, 3]) [06:44] [ComfyUI] [comfy-env] PRE-SERIALIZE 'sky_mask' shape: torch.Size([1, 2002, 1330]) [06:44] [ComfyUI] [comfy-env] PRE-SERIALIZE 'source_image' shape: torch.Size([1, 2002, 1330, 3]) [06:44] [ComfyUI] [SubprocessWorker] serializing kwargs to shm... [06:44] [ComfyUI] [SubprocessWorker] created 0 shm blocks [06:44] [ComfyUI] [SubprocessWorker] sending request via socket... [06:44] [ComfyUI] [SubprocessWorker] call_id=25 sending call_method [06:44] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] _from_shm got dict with keys: ['__type__', 'strategy', 'parent_pid', 'fd', 'storage_size'] [06:44] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZE TensorRef: tensor_size=[1, 2002, 1330, 3] [06:44] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZED tensor shape: torch.Size([1, 2002, 1330, 3]) [06:44] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] _from_shm got dict with keys: ['__type__', 'strategy', 'parent_pid', 'fd', 'storage_size'] [06:44] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZE TensorRef: tensor_size=[1, 2002, 1330, 3] [06:44] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZED tensor shape: torch.Size([1, 2002, 1330, 3]) [06:44] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] _from_shm got dict with keys: ['__type__', 'strategy', 'parent_pid', 'fd', 'storage_size'] [06:44] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZE TensorRef: tensor_size=[1, 2002, 1330] [06:44] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZED tensor shape: torch.Size([1, 2002, 1330]) [06:44] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] _from_shm got dict with keys: ['__type__', 'strategy', 'parent_pid', 'fd', 'storage_size'] [06:44] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZE TensorRef: tensor_size=[1, 2002, 1330, 3] [06:44] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZED tensor shape: torch.Size([1, 2002, 1330, 3]) [06:44] [capture-loop] iter=100 t=12.7s state={'complete': False, 'error': None, 'executedCount': 1, 'wsState': 1} eval_ms=3 ram=4.4/31.2GB (14%) swap=0.5/32.0GB (2%) cpu=6% browser_rss=0.95GB [06:44] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] +538MB (now 538MB) reserved=704MB peak=538MB [06:44] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] WARNING: pyfqmr not installed, skipping decimation. Install with: pip install pyfqmr [06:45] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] -457MB (now 81MB) reserved=704MB peak=538MB [06:49] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] WARNING: pygltflib is required for material post-processing. Install with: pip install pygltflib. Mesh may appear dark or single-sided. [06:49] [ComfyUI] [SubprocessWorker] got response: ok [06:49] [ComfyUI] [comfy-env] [VRAM] Before offload 'DepthAnything3Net_1' (783 MB) -> cpu: 1375 / 24124 MB [06:49] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ping (timeout=5.0s)... [06:49] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ok [06:49] [ComfyUI] [SubprocessWorker] call_id=26 sending model_to_device [06:49] [ComfyUI] Model 'DepthAnything3Net_1' gone from worker (restarted?), treating as offloaded [06:49] [ComfyUI] [comfy-env] [VRAM] After offload 'DepthAnything3Net_1' (783 MB): 1375 / 24124 MB [06:49] [ComfyUI] [comfy-env] DA3_ToMesh done (8.21s): result: NodeOutput [06:49] [ComfyUI] [VRAM] After DA3_ToMesh: 1375 / 24124 MB [06:49] [ComfyUI] Prompt executed in 16.92 seconds [06:49] Node executed (2 total), capturing... [06:49] Frame 5 saved (frame_004.jpg, t=17.5s) [06:49] [capture-node] freeze=4ms shot=416ms unfreeze=13ms saved=True [06:50] Node executed (3 total), capturing... [06:53] Frame 6 saved (frame_005.jpg, t=18.3s) [06:53] [capture-node] freeze=159ms shot=3296ms unfreeze=0ms saved=True [06:53] Execution complete (t=18.3s) [06:53] [3d-wait] Phase 1: Waiting for 'Loading 3D Model' text to disappear... [07:53] [3d-wait] Phase 1 exception after 60.0s: Page.wait_for_function: Timeout 60000ms exceeded. [07:53] [3d-wait] Phase 2: iframes=0 [07:53] [3d-wait] Total 3D wait: 60.0s [07:53] Captured 6 unique frames over 82.1s [07:53] [debug] iframes: 0 [07:53] [timing] trigger_3d_previews: 0.0s [07:56] Saved high-quality screenshot: bas_relief_executed.png [07:56] [timing] final_screenshot: 0.5s [07:56] Capturing execution frames: da3_streaming.json [08:04] Frame 1 saved (frame_000.jpg, t=0.0s) [08:04] Validating workflow... [08:04] [ComfyUI] === /validate endpoint called === [08:04] [ComfyUI] Received JSON with keys: ['prompt'] [08:04] Queuing workflow for execution... [08:04] [ComfyUI] Prompt has 5 nodes [08:04] [ComfyUI] Node 2: DownloadAndLoadDepthAnythingV3Model [08:04] [ComfyUI] Inputs: {"model": "da3_large.safetensors", "precision": "auto", "attention": "auto"} [08:04] [ComfyUI] Node 3: LoadVideo [08:04] [ComfyUI] Inputs: {"file": "robot_unitree.mp4", "video-preview": ""} [08:04] [ComfyUI] Node 4: SaveVideo [08:04] [ComfyUI] Inputs: {"filename_prefix": "video/preview_da3", "format": "auto", "codec": "auto", "video": ["6", 0]} [08:04] [ComfyUI] [ComfyUI-DepthAnythingV3] python: /workspaces/DepthAnythingV3-0759/ComfyUI/.ce/.pixi/envs/depthanythingv3-nodes/bin/python [08:04] [ComfyUI] Node 5: LoadSALADModel [08:04] [ComfyUI] [ComfyUI-DepthAnythingV3] sys_path sent to worker: ['/workspaces/DepthAnythingV3-0759/ComfyUI', '/workspaces/DepthAnythingV3-0759/ComfyUI/custom_nodes/ComfyUI-DepthAnythingV3', '/workspaces/DepthAnythingV3-0759/ComfyUI/.ce/.pixi/envs/depthanythingv3-nodes/lib/python3.1/site-packages', '/workspaces/DepthAnythingV3-0759/ComfyUI/custom_nodes/ComfyUI-DepthAnythingV3'] [08:04] [ComfyUI] Inputs: {} [08:04] [ComfyUI] [SubprocessWorker] is_pixi=True, python=/workspaces/DepthAnythingV3-0759/ComfyUI/.ce/.pixi/envs/depthanythingv3-nodes/bin/python [08:04] [ComfyUI] [SubprocessWorker] launching cmd=['/root/.pixi/bin/pixi', 'run', '--as-is', '--manifest-path', '/workspaces/DepthAnythingV3-0759/ComfyUI/.ce/pixi.toml', '-e', 'depthanythingv3-nodes', 'python', '/tmp/comfyui_pvenv_h_nkff2u/persistent_worker.py', 'abstract://comfy_worker_a8c6282813d7']... [08:04] [ComfyUI] Node 6: DepthAnythingV3_Streaming [08:04] [ComfyUI] Inputs: {"normalization_mode": "Standard", "chunk_size": 30, "overlap": 8, "align_lib": "auto", "align_method": "sim3", "resize_method": "resize", "invert_depth": true, "save_pointcloud": false, "sample_ratio [08:04] [ComfyUI] [SubprocessWorker] socket_addr=abstract://comfy_worker_a8c6282813d7 (abstract, no filesystem path) [08:04] [ComfyUI] Generated prompt_id: 296f598e-8e9b-419a-8113-00d6be9c16af [08:04] [ComfyUI] Calling execution.validate_prompt()... [08:04] [ComfyUI] Validation result: valid=True [08:04] [ComfyUI] === Validation PASSED === [08:04] [ComfyUI] got prompt [08:04] [ComfyUI] [comfy-env] Running LoadSALADModel: [08:04] [ComfyUI] [VRAM] Before LoadSALADModel: 1375 / 24124 MB [08:04] [ComfyUI] [SubprocessWorker] call_method: nodes.load_model.LoadSALADModel.EXECUTE_NORMALIZED [08:04] [ComfyUI] [SubprocessWorker] acquired lock, ensuring started... [08:04] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ping (timeout=5.0s)... [08:04] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: failed [08:04] [ComfyUI] [ComfyUI-DepthAnythingV3] Socket unhealthy, restarting worker... [08:07] Frame 2 saved (frame_001.jpg, t=2.4s) [08:07] [capture-periodic] freeze=3ms shot=202ms unfreeze=7ms saved=True [08:07] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] WARNING: WARNING: You need pytorch with cu130 or higher to use optimized CUDA operations. [08:08] [ComfyUI] [SubprocessWorker] worker started/confirmed [08:08] [ComfyUI] [SubprocessWorker] sending request via socket... [08:08] [ComfyUI] [SubprocessWorker] call_id=27 sending call_method [08:08] [ComfyUI] Warning: You are sending unauthenticated requests to the HF Hub. Please set a HF_TOKEN to enable higher rate limits and faster downloads. [08:08] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] WARNING: Warning: You are sending unauthenticated requests to the HF Hub. Please set a HF_TOKEN to enable higher rate limits and faster downloads. [08:08] [ComfyUI] [08:09] Frame 3 saved (frame_002.jpg, t=4.8s) [08:09] [capture-periodic] freeze=3ms shot=191ms unfreeze=6ms saved=True [08:10] [capture-loop] iter=50 t=6.2s state={'complete': False, 'error': None, 'executedCount': 0, 'wsState': 1} eval_ms=4 ram=5.0/31.2GB (16%) swap=0.5/32.0GB (2%) cpu=7% browser_rss=1.43GB [08:13] [ComfyUI] dino_salad.safetensors: 0%| | 0.00/352M [00:00 [09:05] [ComfyUI] [09:05] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] === END === [09:05] [ComfyUI] [09:05] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] +701MB (now 3337MB) reserved=4320MB peak=4556MB [09:06] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] +350MB (now 3688MB) reserved=4672MB peak=4556MB [09:07] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] +568MB (now 4257MB) reserved=4480MB peak=4556MB [09:09] [capture-loop] iter=550 t=64.6s state={'complete': False, 'error': None, 'executedCount': 0, 'wsState': 1} eval_ms=3 ram=8.2/31.2GB (26%) swap=0.5/32.0GB (2%) cpu=4% browser_rss=0.96GB [09:09] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] +206MB (now 4463MB) reserved=5600MB peak=4556MB [09:10] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] -2947MB (now 1517MB) reserved=7200MB peak=4556MB [09:10] [ComfyUI] [SubprocessWorker] callback 'report_progress' from call_id=29 [09:11] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] +1118MB (now 2635MB) reserved=4160MB peak=4556MB [09:12] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] +701MB (now 3337MB) reserved=4320MB peak=4556MB [09:13] Frame 10 saved (frame_009.jpg, t=68.3s) [09:13] [capture-periodic] freeze=3ms shot=184ms unfreeze=6ms saved=True [09:13] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] +350MB (now 3688MB) reserved=4672MB peak=4556MB [09:14] [capture-loop] iter=600 t=70.3s state={'complete': False, 'error': None, 'executedCount': 0, 'wsState': 1} eval_ms=3 ram=8.4/31.2GB (27%) swap=0.5/32.0GB (2%) cpu=4% browser_rss=0.97GB [09:15] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] +568MB (now 4257MB) reserved=4480MB peak=4556MB [09:17] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] +206MB (now 4463MB) reserved=5856MB peak=4556MB [09:18] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] -2947MB (now 1517MB) reserved=7296MB peak=4556MB [09:18] [ComfyUI] [SubprocessWorker] callback 'report_progress' from call_id=29 [09:19] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] +850MB (now 2367MB) reserved=3328MB peak=4556MB [09:19] Frame 11 saved (frame_010.jpg, t=75.0s) [09:19] [capture-periodic] freeze=3ms shot=185ms unfreeze=6ms saved=True [09:20] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] +467MB (now 2834MB) reserved=3776MB peak=4556MB [09:20] [capture-loop] iter=650 t=76.3s state={'complete': False, 'error': None, 'executedCount': 0, 'wsState': 1} eval_ms=3 ram=8.6/31.2GB (28%) swap=0.5/32.0GB (2%) cpu=5% browser_rss=0.97GB [09:21] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] +514MB (now 3349MB) reserved=3840MB peak=4556MB [09:23] [ComfyUI] [SubprocessWorker] callback 'report_progress' from call_id=29 [09:23] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] -2222MB (now 1128MB) reserved=1184MB peak=4556MB [09:24] Frame 12 saved (frame_011.jpg, t=79.5s) [09:24] [capture-periodic] freeze=3ms shot=191ms unfreeze=6ms saved=True [09:24] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] +214MB (now 1342MB) reserved=1408MB peak=4556MB [09:26] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] -215MB (now 1128MB) reserved=1632MB peak=4556MB [09:26] [capture-loop] iter=700 t=82.3s state={'complete': False, 'error': None, 'executedCount': 0, 'wsState': 1} eval_ms=4 ram=9.5/31.2GB (31%) swap=0.5/32.0GB (2%) cpu=5% browser_rss=0.97GB [09:26] [ComfyUI] [SubprocessWorker] callback 'report_progress' from call_id=29 [09:27] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] +221MB (now 1350MB) reserved=1728MB peak=4556MB [09:28] [ComfyUI] [SubprocessWorker] callback 'report_progress' from call_id=29 [09:28] Frame 13 saved (frame_012.jpg, t=84.0s) [09:28] [capture-periodic] freeze=3ms shot=185ms unfreeze=8ms saved=True [09:28] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] -222MB (now 1128MB) reserved=1760MB peak=4556MB [09:30] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] +211MB (now 1339MB) reserved=1760MB peak=4556MB [09:30] [ComfyUI] [SubprocessWorker] callback 'report_progress' from call_id=29 [09:31] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] -212MB (now 1128MB) reserved=1760MB peak=4556MB [09:31] Frame 14 saved (frame_013.jpg, t=86.3s) [09:31] [capture-periodic] freeze=8ms shot=183ms unfreeze=3ms saved=True [09:32] [ComfyUI] [SubprocessWorker] callback 'report_progress' from call_id=29 [09:32] [capture-loop] iter=750 t=88.2s state={'complete': False, 'error': None, 'executedCount': 0, 'wsState': 1} eval_ms=4 ram=9.3/31.2GB (30%) swap=0.5/32.0GB (2%) cpu=6% browser_rss=0.97GB [09:33] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] +319MB (now 1448MB) reserved=1760MB peak=4556MB [09:33] Frame 15 saved (frame_014.jpg, t=88.6s) [09:33] [capture-periodic] freeze=3ms shot=200ms unfreeze=7ms saved=True [09:33] [ComfyUI] [SubprocessWorker] callback 'report_progress' from call_id=29 [09:34] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] -320MB (now 1128MB) reserved=1760MB peak=4556MB [09:35] [ComfyUI] [SubprocessWorker] callback 'report_progress' from call_id=29 [09:35] Frame 16 saved (frame_015.jpg, t=90.9s) [09:35] [capture-periodic] freeze=3ms shot=191ms unfreeze=6ms saved=True [09:36] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] +218MB (now 1346MB) reserved=1760MB peak=4556MB [09:37] [ComfyUI] [SubprocessWorker] callback 'report_progress' from call_id=29 [09:37] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] WARNING: faiss not installed, skipping loop closure. pip install faiss-cpu [09:37] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] -219MB (now 1128MB) reserved=1760MB peak=4556MB [09:38] Frame 17 saved (frame_016.jpg, t=93.2s) [09:38] [capture-periodic] freeze=3ms shot=196ms unfreeze=6ms saved=True [09:38] [capture-loop] iter=800 t=94.4s state={'complete': False, 'error': None, 'executedCount': 0, 'wsState': 1} eval_ms=4 ram=10.5/31.2GB (34%) swap=0.5/32.0GB (2%) cpu=5% browser_rss=0.97GB [09:41] [ComfyUI] [SubprocessWorker] callback 'report_progress' from call_id=29 [09:42] Frame 18 saved (frame_017.jpg, t=97.8s) [09:42] [capture-periodic] freeze=3ms shot=147ms unfreeze=8ms saved=True [09:44] [capture-loop] iter=850 t=100.1s state={'complete': False, 'error': None, 'executedCount': 0, 'wsState': 1} eval_ms=3 ram=8.6/31.2GB (28%) swap=0.5/32.0GB (2%) cpu=5% browser_rss=0.97GB [09:50] [capture-loop] iter=900 t=106.1s state={'complete': False, 'error': None, 'executedCount': 0, 'wsState': 1} eval_ms=3 ram=8.6/31.2GB (28%) swap=0.5/32.0GB (2%) cpu=4% browser_rss=0.97GB [09:56] [capture-loop] iter=950 t=111.9s state={'complete': False, 'error': None, 'executedCount': 0, 'wsState': 1} eval_ms=3 ram=8.6/31.2GB (28%) swap=0.5/32.0GB (2%) cpu=4% browser_rss=0.97GB [10:02] [capture-loop] iter=1000 t=117.6s state={'complete': False, 'error': None, 'executedCount': 0, 'wsState': 1} eval_ms=3 ram=8.7/31.2GB (28%) swap=0.5/32.0GB (2%) cpu=5% browser_rss=0.97GB [10:05] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [10:05] [ComfyUI] === WATCHDOG TICK 2 === [10:05] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] Thread 0x00007d37796a1640 (most recent call first): [10:05] [ComfyUI] File "/workspaces/DepthAnythingV3-0759/ComfyUI/.ce/.pixi/envs/depthanythingv3-nodes/lib/python3.12/threading.py", line 359 in wait [10:05] [ComfyUI] File "/workspaces/DepthAnythingV3-0759/ComfyUI/.ce/.pixi/envs/depthanythingv3-nodes/lib/python3.12/threading.py", line 655 in wait [10:05] [ComfyUI] File "/workspaces/DepthAnythingV3-0759/ComfyUI/.ce/.pixi/envs/depthanythingv3-nodes/lib/python3.12/site-packages/tqdm/_monitor.py", line 60 in run [10:05] [ComfyUI] File "/workspaces/DepthAnythingV3-0759/ComfyUI/.ce/.pixi/envs/depthanythingv3-nodes/lib/python3.12/threading.py", line 1075 in _bootstrap_inner [10:05] [ComfyUI] File "/workspaces/DepthAnythingV3-0759/ComfyUI/.ce/.pixi/envs/depthanythingv3-nodes/lib/python3.12/threading.py", line 1032 in _bootstrap [10:05] [ComfyUI] [10:05] [ComfyUI] Thread 0x00007d39328a9640 (most recent call first): [10:05] [ComfyUI] File "/tmp/comfyui_pvenv_h_nkff2u/persistent_worker.py", line 106 in _vram_poller [10:05] [ComfyUI] File "/workspaces/DepthAnythingV3-0759/ComfyUI/.ce/.pixi/envs/depthanythingv3-nodes/lib/python3.12/threading.py", line 1012 in run [10:05] [ComfyUI] File "/workspaces/DepthAnythingV3-0759/ComfyUI/.ce/.pixi/envs/depthanythingv3-nodes/lib/python3.12/threading.py", line 1075 in _bootstrap_inner [10:05] [ComfyUI] File "/workspaces/DepthAnythingV3-0759/ComfyUI/.ce/.pixi/envs/depthanythingv3-nodes/lib/python3.12/threading.py", line 1032 in _bootstrap [10:05] [ComfyUI] [10:05] [ComfyUI] Current thread 0x00007d39330ae640 (most recent call first): [10:05] [ComfyUI] File "/tmp/comfyui_pvenv_h_nkff2u/persistent_worker.py", line 52 in _watchdog [10:05] [ComfyUI] File "/workspaces/DepthAnythingV3-0759/ComfyUI/.ce/.pixi/envs/depthanythingv3-nodes/lib/python3.12/threading.py", line 1012 in run [10:05] [ComfyUI] File "/workspaces/DepthAnythingV3-0759/ComfyUI/.ce/.pixi/envs/depthanythingv3-nodes/lib/python3.12/threading.py", line 1075 in _bootstrap_inner [10:05] [ComfyUI] File "/workspaces/DepthAnythingV3-0759/ComfyUI/.ce/.pixi/envs/depthanythingv3-nodes/lib/python3.12/threading.py", line 1032 in _bootstrap [10:05] [ComfyUI] [10:05] [ComfyUI] Thread 0x00007d39336a9740 (most recent call first): [10:05] [ComfyUI] File "/workspaces/DepthAnythingV3-0759/ComfyUI/.ce/.pixi/envs/depthanythingv3-nodes/lib/python3.12/zipfile/__init__.py", line 1235 in write [10:05] [ComfyUI] File "/workspaces/DepthAnythingV3-0759/ComfyUI/.ce/.pixi/envs/depthanythingv3-nodes/lib/python3.12/site-packages/numpy/lib/_format_impl.py", line 777 in write_array [10:05] [ComfyUI] File "/workspaces/DepthAnythingV3-0759/ComfyUI/.ce/.pixi/envs/depthanythingv3-nodes/lib/python3.12/site-packages/numpy/lib/_npyio_impl.py", line 786 in _savez [10:05] [ComfyUI] File "/workspaces/DepthAnythingV3-0759/ComfyUI/.ce/.pixi/envs/depthanythingv3-nodes/lib/python3.12/site-packages/numpy/lib/_npyio_impl.py", line 753 in savez_compressed [10:05] [ComfyUI] File "/workspaces/DepthAnythingV3-0759/ComfyUI/custom_nodes/ComfyUI-DepthAnythingV3/nodes/streaming/node.py", line 271 in execute [10:05] [ComfyUI] File "/workspaces/DepthAnythingV3-0759/ComfyUI/comfy_api/latest/_io.py", line 1833 in EXECUTE_NORMALIZED [10:05] [ComfyUI] File "/tmp/comfyui_pvenv_h_nkff2u/persistent_worker.py", line 1762 in main [10:05] [ComfyUI] File "/tmp/comfyui_pvenv_h_nkff2u/persistent_worker.py", line 1807 in [10:05] [ComfyUI] [10:05] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] === END === [10:05] [ComfyUI] [10:08] [capture-loop] iter=1050 t=123.5s state={'complete': False, 'error': None, 'executedCount': 0, 'wsState': 1} eval_ms=3 ram=8.6/31.2GB (28%) swap=0.5/32.0GB (2%) cpu=4% browser_rss=0.97GB [10:13] [capture-loop] iter=1100 t=129.2s state={'complete': False, 'error': None, 'executedCount': 0, 'wsState': 1} eval_ms=3 ram=8.6/31.2GB (28%) swap=0.5/32.0GB (2%) cpu=4% browser_rss=0.97GB [10:19] [capture-loop] iter=1150 t=135.0s state={'complete': False, 'error': None, 'executedCount': 0, 'wsState': 1} eval_ms=3 ram=8.7/31.2GB (28%) swap=0.5/32.0GB (2%) cpu=5% browser_rss=0.97GB [10:25] [capture-loop] iter=1200 t=140.7s state={'complete': False, 'error': None, 'executedCount': 0, 'wsState': 1} eval_ms=4 ram=10.8/31.2GB (34%) swap=0.5/32.0GB (2%) cpu=4% browser_rss=0.97GB [10:29] [ComfyUI] [SubprocessWorker] got response: ok [10:31] [capture-loop] iter=1250 t=146.6s state={'complete': False, 'error': None, 'executedCount': 0, 'wsState': 1} eval_ms=4 ram=10.7/31.2GB (34%) swap=0.5/32.0GB (2%) cpu=4% browser_rss=0.97GB [10:31] [ComfyUI] [comfy-env] Created 2 model patchers: ['DepthAnything3Net_1', 'VPRModel_2'] [10:31] [ComfyUI] [comfy-env] DepthAnythingV3_Streaming done (138.18s): result: NodeOutput [10:31] [ComfyUI] [VRAM] After DepthAnythingV3_Streaming: 2443 / 24124 MB [10:31] [ComfyUI] [VRAM] Loaded: SubprocessModel (335 MB), SubprocessModel (783 MB) [10:32] Frame 19 saved (frame_018.jpg, t=147.5s) [10:32] [capture-periodic] freeze=3ms shot=173ms unfreeze=3ms saved=True [10:35] [ComfyUI] Prompt executed in 151.01 seconds [10:35] Node executed (1 total), capturing... [10:36] Frame 20 saved (frame_019.jpg, t=151.5s) [10:36] [capture-node] freeze=3ms shot=188ms unfreeze=17ms saved=True [10:36] Execution complete (t=151.9s) [10:36] Captured 20 unique frames over 151.92s [10:36] [debug] iframes: 0 [10:36] [timing] trigger_3d_previews: 0.0s [10:38] Saved high-quality screenshot: da3_streaming_executed.png [10:38] [timing] final_screenshot: 0.4s [10:38] [ComfyUI] [comfy-env] [VRAM] Before offload 'VPRModel_2' (335 MB) -> cpu: 2443 / 24124 MB [10:38] Capturing execution frames: simple.json [10:38] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ping (timeout=5.0s)... [10:39] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ok [10:39] [ComfyUI] [SubprocessWorker] call_id=30 sending model_to_device [10:39] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] -237MB (now 892MB) reserved=1760MB peak=4556MB [10:39] [ComfyUI] [comfy-env] [VRAM] After offload 'VPRModel_2' (335 MB): 1547 / 24124 MB [10:39] [ComfyUI] [comfy-env] [VRAM] Before offload 'DepthAnything3Net_1' (783 MB) -> cpu: 1547 / 24124 MB [10:39] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ping (timeout=5.0s)... [10:39] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ok [10:39] [ComfyUI] [SubprocessWorker] call_id=31 sending model_to_device [10:39] [ComfyUI] [comfy-env] [VRAM] After offload 'DepthAnything3Net_1' (783 MB): 747 / 24124 MB [10:45] Frame 1 saved (frame_000.jpg, t=0.0s) [10:46] Validating workflow... [10:46] [ComfyUI] === /validate endpoint called === [10:46] [ComfyUI] Received JSON with keys: ['prompt'] [10:46] Queuing workflow for execution... [10:46] [ComfyUI] Prompt has 4 nodes [10:46] [ComfyUI] Node 2: LoadImage [10:46] [ComfyUI] Inputs: {"image": "bridge.jpg"} [10:46] [ComfyUI] Node 3: DownloadAndLoadDepthAnythingV3Model [10:46] [ComfyUI] Inputs: {"model": "da3_large.safetensors", "precision": "auto", "attention": "auto"} [10:46] [ComfyUI] Node 4: PreviewImage [10:46] [ComfyUI] Inputs: {"images": ["14", 0]} [10:46] [ComfyUI] Node 14: DepthAnything_V3 [10:46] [ComfyUI] Inputs: {"normalization_mode": "V2-Style", "resize_method": "resize", "invert_depth": true, "keep_model_size": false, "da3_model": ["3", 0], "images": ["2", 0]} [10:46] [ComfyUI] Generated prompt_id: dc606192-af09-4620-ad51-156ef5efb8ca [10:46] [ComfyUI] Calling execution.validate_prompt()... [10:46] [ComfyUI] Validation result: valid=True [10:46] [ComfyUI] === Validation PASSED === [10:46] [ComfyUI] got prompt [10:46] [ComfyUI] [comfy-env] Running DownloadAndLoadDepthAnythingV3Model: model: 'da3_large.safetensors', precision: 'auto', attention: 'auto' [10:46] [ComfyUI] [VRAM] Before DownloadAndLoadDepthAnythingV3Model: 747 / 24124 MB [10:46] [ComfyUI] [SubprocessWorker] call_method: nodes.load_model.DownloadAndLoadDepthAnythingV3Model.EXECUTE_NORMALIZED [10:46] [ComfyUI] [ComfyUI-DepthAnythingV3] python: /workspaces/DepthAnythingV3-0759/ComfyUI/.ce/.pixi/envs/depthanythingv3-nodes/bin/python [10:46] [ComfyUI] [SubprocessWorker] acquired lock, ensuring started... [10:46] [ComfyUI] [ComfyUI-DepthAnythingV3] sys_path sent to worker: ['/workspaces/DepthAnythingV3-0759/ComfyUI', '/workspaces/DepthAnythingV3-0759/ComfyUI/custom_nodes/ComfyUI-DepthAnythingV3', '/workspaces/DepthAnythingV3-0759/ComfyUI/.ce/.pixi/envs/depthanythingv3-nodes/lib/python3.1/site-packages', '/workspaces/DepthAnythingV3-0759/ComfyUI/custom_nodes/ComfyUI-DepthAnythingV3'] [10:46] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ping (timeout=5.0s)... [10:46] [ComfyUI] [SubprocessWorker] is_pixi=True, python=/workspaces/DepthAnythingV3-0759/ComfyUI/.ce/.pixi/envs/depthanythingv3-nodes/bin/python [10:46] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: failed [10:46] [ComfyUI] [SubprocessWorker] launching cmd=['/root/.pixi/bin/pixi', 'run', '--as-is', '--manifest-path', '/workspaces/DepthAnythingV3-0759/ComfyUI/.ce/pixi.toml', '-e', 'depthanythingv3-nodes', 'python', '/tmp/comfyui_pvenv_h_nkff2u/persistent_worker.py', 'abstract://comfy_worker_5d521a502e82']... [10:46] [ComfyUI] [ComfyUI-DepthAnythingV3] Socket unhealthy, restarting worker... [10:46] [ComfyUI] [SubprocessWorker] socket_addr=abstract://comfy_worker_5d521a502e82 (abstract, no filesystem path) [10:46] [ComfyUI] [comfy-env] Invalidated 2 stale model patchers (will be cleaned up during next unload) [10:48] Frame 2 saved (frame_001.jpg, t=2.5s) [10:48] [capture-periodic] freeze=3ms shot=214ms unfreeze=3ms saved=True [10:49] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] WARNING: WARNING: You need pytorch with cu130 or higher to use optimized CUDA operations. [10:49] [ComfyUI] [SubprocessWorker] worker started/confirmed [10:49] [ComfyUI] [SubprocessWorker] serializing kwargs to shm... [10:49] [ComfyUI] [SubprocessWorker] created 0 shm blocks [10:49] [ComfyUI] [SubprocessWorker] sending request via socket... [10:49] [ComfyUI] [SubprocessWorker] call_id=32 sending call_method [10:49] [ComfyUI] [SubprocessWorker] got response: ok [10:49] [ComfyUI] [comfy-env] [VRAM] Before offload 'VPRModel_2' (335 MB) -> cpu: 563 / 24124 MB [10:49] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ping (timeout=5.0s)... [10:49] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ok [10:49] [ComfyUI] [SubprocessWorker] call_id=33 sending model_to_device [10:49] [ComfyUI] Model 'VPRModel_2' gone from worker (restarted?), treating as offloaded [10:49] [ComfyUI] [comfy-env] [VRAM] After offload 'VPRModel_2' (335 MB): 563 / 24124 MB [10:49] [ComfyUI] [comfy-env] [VRAM] Before offload 'DepthAnything3Net_1' (783 MB) -> cpu: 563 / 24124 MB [10:49] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ping (timeout=5.0s)... [10:49] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ok [10:49] [ComfyUI] [SubprocessWorker] call_id=34 sending model_to_device [10:49] [ComfyUI] Model 'DepthAnything3Net_1' gone from worker (restarted?), treating as offloaded [10:49] [ComfyUI] [comfy-env] [VRAM] After offload 'DepthAnything3Net_1' (783 MB): 563 / 24124 MB [10:49] [ComfyUI] [comfy-env] DownloadAndLoadDepthAnythingV3Model done (3.60s): result: NodeOutput [10:49] [ComfyUI] [VRAM] After DownloadAndLoadDepthAnythingV3Model: 563 / 24124 MB [10:49] [ComfyUI] [comfy-env] Running DepthAnything_V3: normalization_mode: 'V2-Style', resize_method: 'resize', invert_depth: True, keep_model_size: False, da3_model: dict, images: torch.float32 [1,683,1024,3] cpu [10:49] [ComfyUI] [VRAM] Before DepthAnything_V3: 563 / 24124 MB [10:49] [ComfyUI] [SubprocessWorker] call_method: nodes.nodes_inference.DepthAnything_V3.EXECUTE_NORMALIZED [10:49] [ComfyUI] [SubprocessWorker] acquired lock, ensuring started... [10:49] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ping (timeout=5.0s)... [10:49] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ok [10:49] [ComfyUI] [SubprocessWorker] worker started/confirmed [10:49] [ComfyUI] [comfy-env] PRE-SERIALIZE 'images' shape: torch.Size([1, 683, 1024, 3]) [10:49] [ComfyUI] [SubprocessWorker] serializing kwargs to shm... [10:49] [ComfyUI] [SubprocessWorker] created 0 shm blocks [10:49] [ComfyUI] [SubprocessWorker] sending request via socket... [10:49] [ComfyUI] [SubprocessWorker] call_id=35 sending call_method [10:49] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] _from_shm got dict with keys: ['__type__', 'strategy', 'parent_pid', 'fd', 'storage_size'] [10:49] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZE TensorRef: tensor_size=[1, 683, 1024, 3] [10:49] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZED tensor shape: torch.Size([1, 683, 1024, 3]) [10:50] Frame 3 saved (frame_002.jpg, t=4.8s) [10:50] [capture-periodic] freeze=3ms shot=209ms unfreeze=3ms saved=True [10:51] [ComfyUI] [SubprocessWorker] callback 'request_vram_budget' from call_id=35 [10:51] [ComfyUI] [comfy-env] VRAM request: 0.82GB, free before eviction: 24.71GB [10:51] [ComfyUI] [comfy-env] VRAM after eviction: 24.71GB [10:51] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] +426MB (now 426MB) reserved=448MB peak=426MB [10:51] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] WARNING: WARNING: This model does not support sky segmentation. V2-Style normalization will work but without sky masking. Use Mono/Metric/Nested models for best V2-Style results. [10:51] [capture-loop] iter=50 t=6.2s state={'complete': False, 'error': None, 'executedCount': 0, 'wsState': 1} eval_ms=4 ram=4.7/31.2GB (15%) swap=0.5/32.0GB (2%) cpu=8% browser_rss=1.08GB [10:52] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] +426MB (now 853MB) reserved=864MB peak=853MB [10:53] [ComfyUI] [SubprocessWorker] callback 'report_progress' from call_id=35 [10:53] [ComfyUI] [SubprocessWorker] got response: ok [10:53] [ComfyUI] [comfy-env] Created 1 model patchers: ['DepthAnything3Net_1'] [10:53] [ComfyUI] [comfy-env] DepthAnything_V3 done (4.01s): result: NodeOutput [10:53] [ComfyUI] [VRAM] After DepthAnything_V3: 3239 / 24124 MB [10:53] [ComfyUI] [VRAM] Loaded: SubprocessModel (783 MB) [10:53] [ComfyUI] Prompt executed in 7.73 seconds [10:53] [ComfyUI] [comfy-env] [VRAM] Before offload 'DepthAnything3Net_1' (783 MB) -> cpu: 3239 / 24124 MB [10:53] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ping (timeout=5.0s)... [10:53] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ok [10:53] [ComfyUI] [SubprocessWorker] call_id=36 sending model_to_device [10:53] Node executed (1 total), capturing... [10:53] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] -230MB (now 623MB) reserved=2592MB peak=1172MB [10:54] Frame 4 saved (frame_003.jpg, t=8.3s) [10:54] [capture-node] freeze=3ms shot=219ms unfreeze=0ms saved=True [10:54] Execution complete (t=8.3s) [10:54] Captured 4 unique frames over 8.64s [10:54] [debug] iframes: 0 [10:54] [timing] trigger_3d_previews: 0.0s [10:54] [ComfyUI] [comfy-env] [VRAM] After offload 'DepthAnything3Net_1' (783 MB): 711 / 24124 MB [10:56] Saved high-quality screenshot: simple_executed.png [10:56] [timing] final_screenshot: 0.5s [10:57] Capturing execution frames: video_multiview_depth.json [11:04] Frame 1 saved (frame_000.jpg, t=0.0s) [11:04] Validating workflow... [11:04] [ComfyUI] === /validate endpoint called === [11:04] [ComfyUI] Received JSON with keys: ['prompt'] [11:04] Queuing workflow for execution... [11:04] [ComfyUI] Prompt has 9 nodes [11:04] [ComfyUI] Node 3: DownloadAndLoadDepthAnythingV3Model [11:04] [ComfyUI] Inputs: {"model": "da3_base.safetensors", "precision": "auto", "attention": "auto"} [11:04] [ComfyUI] Node 52: CreateVideo [11:04] [ComfyUI] Inputs: {"fps": 10, "images": ["62", 0]} [11:04] [ComfyUI] Node 53: CreateVideo [11:04] [ComfyUI] Inputs: {"fps": 10, "images": ["61", 0]} [11:04] [ComfyUI] Node 54: SaveVideo [11:04] [ComfyUI] Inputs: {"filename_prefix": "video/ComfyUI", "format": "auto", "codec": "auto", "video": ["52", 0]} [11:04] [ComfyUI] Node 55: SaveVideo [11:04] [ComfyUI] Inputs: {"filename_prefix": "video/ComfyUI", "format": "auto", "codec": "auto", "video": ["53", 0]} [11:04] [ComfyUI] Node 61: DepthAnything_V3 [11:04] [ComfyUI] Inputs: {"normalization_mode": "V2-Style", "resize_method": "resize", "invert_depth": false, "keep_model_size": false, "da3_model": ["3", 0], "images": ["71", 0]} [11:04] [ComfyUI] Node 62: DepthAnythingV3_MultiView [11:04] [ComfyUI] Inputs: {"normalization_mode": "V2-Style", "resize_method": "resize", "invert_depth": false, "keep_model_size": false, "da3_model": ["3", 0], "images": ["71", 0]} [11:04] [ComfyUI] Node 66: PreviewImage [11:04] [ComfyUI] Inputs: {"images": ["71", 0]} [11:04] [ComfyUI] Node 71: LoadImagesFromFolderKJ [11:04] [ComfyUI] Inputs: {"folder": "input/robotframes", "width": 1024, "height": 576, "keep_aspect_ratio": "crop", "image_load_cap": 0, "start_index": 0, "include_subfolders": false} [11:04] [ComfyUI] Generated prompt_id: 381b67a8-a83c-4029-9cee-c489632b8201 [11:04] [ComfyUI] Calling execution.validate_prompt()... [11:04] [ComfyUI] Validation result: valid=True [11:04] [ComfyUI] === Validation PASSED === [11:04] [ComfyUI] got prompt [11:05] [ComfyUI] [comfy-env] Running DownloadAndLoadDepthAnythingV3Model: model: 'da3_base.safetensors', precision: 'auto', attention: 'auto' [11:05] [ComfyUI] [ComfyUI-DepthAnythingV3] python: /workspaces/DepthAnythingV3-0759/ComfyUI/.ce/.pixi/envs/depthanythingv3-nodes/bin/python [11:05] [ComfyUI] [VRAM] Before DownloadAndLoadDepthAnythingV3Model: 711 / 24124 MB [11:05] [ComfyUI] [ComfyUI-DepthAnythingV3] sys_path sent to worker: ['/workspaces/DepthAnythingV3-0759/ComfyUI', '/workspaces/DepthAnythingV3-0759/ComfyUI/custom_nodes/ComfyUI-DepthAnythingV3', '/workspaces/DepthAnythingV3-0759/ComfyUI/.ce/.pixi/envs/depthanythingv3-nodes/lib/python3.1/site-packages', '/workspaces/DepthAnythingV3-0759/ComfyUI/custom_nodes/ComfyUI-DepthAnythingV3'] [11:05] [ComfyUI] [SubprocessWorker] call_method: nodes.load_model.DownloadAndLoadDepthAnythingV3Model.EXECUTE_NORMALIZED [11:05] [ComfyUI] [SubprocessWorker] is_pixi=True, python=/workspaces/DepthAnythingV3-0759/ComfyUI/.ce/.pixi/envs/depthanythingv3-nodes/bin/python [11:05] [ComfyUI] [SubprocessWorker] acquired lock, ensuring started... [11:05] [ComfyUI] [SubprocessWorker] launching cmd=['/root/.pixi/bin/pixi', 'run', '--as-is', '--manifest-path', '/workspaces/DepthAnythingV3-0759/ComfyUI/.ce/pixi.toml', '-e', 'depthanythingv3-nodes', 'python', '/tmp/comfyui_pvenv_h_nkff2u/persistent_worker.py', 'abstract://comfy_worker_95f149cae1da']... [11:05] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ping (timeout=5.0s)... [11:05] [ComfyUI] [SubprocessWorker] socket_addr=abstract://comfy_worker_95f149cae1da (abstract, no filesystem path) [11:05] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: failed [11:05] [ComfyUI] [ComfyUI-DepthAnythingV3] Socket unhealthy, restarting worker... [11:05] [ComfyUI] [comfy-env] Invalidated 1 stale model patchers (will be cleaned up during next unload) [11:05] Node executed (1 total), capturing... [11:05] Frame 2 saved (frame_001.jpg, t=1.6s) [11:05] [capture-node] freeze=3ms shot=206ms unfreeze=8ms saved=True [11:08] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] WARNING: WARNING: You need pytorch with cu130 or higher to use optimized CUDA operations. [11:08] [ComfyUI] [SubprocessWorker] worker started/confirmed [11:08] [ComfyUI] [SubprocessWorker] serializing kwargs to shm... [11:08] [ComfyUI] [SubprocessWorker] created 0 shm blocks [11:08] [ComfyUI] [SubprocessWorker] sending request via socket... [11:08] [ComfyUI] [SubprocessWorker] call_id=37 sending call_method [11:09] [ComfyUI] Warning: You are sending unauthenticated requests to the HF Hub. Please set a HF_TOKEN to enable higher rate limits and faster downloads. [11:09] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] WARNING: Warning: You are sending unauthenticated requests to the HF Hub. Please set a HF_TOKEN to enable higher rate limits and faster downloads. [11:09] [ComfyUI] [11:10] Frame 3 saved (frame_002.jpg, t=6.2s) [11:10] [capture-periodic] freeze=3ms shot=202ms unfreeze=7ms saved=True [11:10] [capture-loop] iter=50 t=6.7s state={'complete': False, 'error': None, 'executedCount': 1, 'wsState': 1} eval_ms=4 ram=4.5/31.2GB (14%) swap=0.5/32.0GB (2%) cpu=5% browser_rss=1.01GB [11:12] [ComfyUI] model.safetensors: 0%| | 0.00/542M [00:00 cpu: 563 / 24124 MB [11:15] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ping (timeout=5.0s)... [11:15] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ok [11:15] [ComfyUI] [SubprocessWorker] call_id=38 sending model_to_device [11:15] [ComfyUI] Model 'DepthAnything3Net_1' gone from worker (restarted?), treating as offloaded [11:15] [ComfyUI] [comfy-env] [VRAM] After offload 'DepthAnything3Net_1' (783 MB): 563 / 24124 MB [11:15] [ComfyUI] [comfy-env] DownloadAndLoadDepthAnythingV3Model done (10.13s): result: NodeOutput [11:15] [ComfyUI] [VRAM] After DownloadAndLoadDepthAnythingV3Model: 563 / 24124 MB [11:15] [ComfyUI] [comfy-env] Running DepthAnything_V3: normalization_mode: 'V2-Style', resize_method: 'resize', invert_depth: False, keep_model_size: False, da3_model: dict, images: torch.float32 [13,576,1024,3] cpu [11:15] [ComfyUI] [VRAM] Before DepthAnything_V3: 563 / 24124 MB [11:15] [ComfyUI] [SubprocessWorker] call_method: nodes.nodes_inference.DepthAnything_V3.EXECUTE_NORMALIZED [11:15] [ComfyUI] [SubprocessWorker] acquired lock, ensuring started... [11:15] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ping (timeout=5.0s)... [11:15] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ok [11:15] [ComfyUI] [SubprocessWorker] worker started/confirmed [11:15] [ComfyUI] [comfy-env] PRE-SERIALIZE 'images' shape: torch.Size([13, 576, 1024, 3]) [11:15] [ComfyUI] [SubprocessWorker] serializing kwargs to shm... [11:15] [ComfyUI] [SubprocessWorker] created 0 shm blocks [11:15] [ComfyUI] [SubprocessWorker] sending request via socket... [11:15] [ComfyUI] [SubprocessWorker] call_id=39 sending call_method [11:15] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] _from_shm got dict with keys: ['__type__', 'strategy', 'parent_pid', 'fd', 'storage_size'] [11:15] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZE TensorRef: tensor_size=[13, 576, 1024, 3] [11:15] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZED tensor shape: torch.Size([13, 576, 1024, 3]) [11:16] [capture-loop] iter=100 t=12.5s state={'complete': False, 'error': None, 'executedCount': 1, 'wsState': 1} eval_ms=4 ram=4.8/31.2GB (15%) swap=0.5/32.0GB (2%) cpu=5% browser_rss=1.01GB [11:16] [ComfyUI] [SubprocessWorker] callback 'request_vram_budget' from call_id=39 [11:16] [ComfyUI] [comfy-env] VRAM request: 0.27GB, free before eviction: 24.71GB [11:16] [ComfyUI] [comfy-env] VRAM after eviction: 24.71GB [11:16] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] +253MB (now 253MB) reserved=256MB peak=253MB [11:16] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] WARNING: WARNING: This model does not support sky segmentation. V2-Style normalization will work but without sky masking. Use Mono/Metric/Nested models for best V2-Style results. [11:17] Frame 5 saved (frame_004.jpg, t=13.1s) [11:17] [capture-periodic] freeze=3ms shot=192ms unfreeze=8ms saved=True [11:19] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] +384MB (now 638MB) reserved=896MB peak=638MB [11:19] [ComfyUI] [SubprocessWorker] callback 'report_progress' from call_id=39 [11:19] [ComfyUI] [SubprocessWorker] callback 'report_progress' from call_id=39 [11:19] [ComfyUI] [SubprocessWorker] callback 'report_progress' from call_id=39 [11:19] [ComfyUI] [SubprocessWorker] callback 'report_progress' from call_id=39 [11:20] [ComfyUI] [SubprocessWorker] callback 'report_progress' from call_id=39 [11:20] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] -320MB (now 318MB) reserved=1088MB peak=638MB [11:20] [ComfyUI] [SubprocessWorker] callback 'report_progress' from call_id=39 [11:20] [ComfyUI] [SubprocessWorker] callback 'report_progress' from call_id=39 [11:20] [ComfyUI] [SubprocessWorker] callback 'report_progress' from call_id=39 [11:21] Frame 6 saved (frame_005.jpg, t=17.7s) [11:21] [capture-periodic] freeze=3ms shot=196ms unfreeze=8ms saved=True [11:22] [capture-loop] iter=150 t=18.6s state={'complete': False, 'error': None, 'executedCount': 1, 'wsState': 1} eval_ms=4 ram=6.1/31.2GB (20%) swap=0.5/32.0GB (2%) cpu=5% browser_rss=1.01GB [11:22] [ComfyUI] [SubprocessWorker] got response: ok [11:22] [ComfyUI] [comfy-env] Created 1 model patchers: ['DepthAnything3Net_1'] [11:22] [ComfyUI] [comfy-env] DepthAnything_V3 done (7.14s): result: NodeOutput [11:22] [ComfyUI] [VRAM] After DepthAnything_V3: 1735 / 24124 MB [11:22] [ComfyUI] [VRAM] Loaded: SubprocessModel (258 MB) [11:22] [ComfyUI] [comfy-env] Running DepthAnythingV3_MultiView: normalization_mode: 'V2-Style', resize_method: 'resize', invert_depth: False, keep_model_size: False, da3_model: dict, images: torch.float32 [13,576,1024,3] cpu [11:22] [ComfyUI] [VRAM] Before DepthAnythingV3_MultiView: 1735 / 24124 MB [11:22] [ComfyUI] [VRAM] Loaded: SubprocessModel (258 MB) [11:22] [ComfyUI] [SubprocessWorker] call_method: nodes.nodes_multiview.DepthAnythingV3_MultiView.EXECUTE_NORMALIZED [11:22] [ComfyUI] [SubprocessWorker] acquired lock, ensuring started... [11:22] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ping (timeout=5.0s)... [11:22] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ok [11:22] [ComfyUI] [SubprocessWorker] worker started/confirmed [11:22] [ComfyUI] [comfy-env] PRE-SERIALIZE 'images' shape: torch.Size([13, 576, 1024, 3]) [11:22] [ComfyUI] [SubprocessWorker] serializing kwargs to shm... [11:22] [ComfyUI] [SubprocessWorker] created 0 shm blocks [11:22] [ComfyUI] [SubprocessWorker] sending request via socket... [11:22] [ComfyUI] [SubprocessWorker] call_id=40 sending call_method [11:22] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] _from_shm got dict with keys: ['__type__', 'strategy', 'parent_pid', 'fd', 'storage_size'] [11:22] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZE TensorRef: tensor_size=[13, 576, 1024, 3] [11:22] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [comfy-env] DESERIALIZED tensor shape: torch.Size([13, 576, 1024, 3]) [11:22] [ComfyUI] [SubprocessWorker] callback 'request_vram_budget' from call_id=40 [11:22] [ComfyUI] [comfy-env] VRAM request: 0.27GB, free before eviction: 23.48GB [11:22] [ComfyUI] [comfy-env] VRAM after eviction: 23.48GB [11:22] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] WARNING: WARNING: This model does not support sky segmentation. V2-Style normalization will work but without sky masking. Use Mono/Metric/Nested models for best V2-Style results. [11:22] Node executed (2 total), capturing... [11:23] Frame 7 saved (frame_006.jpg, t=19.3s) [11:23] [capture-node] freeze=4ms shot=205ms unfreeze=4ms saved=True [11:23] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] +392MB (now 711MB) reserved=1312MB peak=711MB [11:24] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] +664MB (now 1376MB) reserved=2624MB peak=1598MB [11:25] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] -723MB (now 653MB) reserved=2560MB peak=1598MB [11:25] Frame 8 saved (frame_007.jpg, t=21.6s) [11:25] [capture-periodic] freeze=3ms shot=212ms unfreeze=8ms saved=True [11:26] [ComfyUI] [worker:ComfyUI-DepthAnythingV3] [VRAM] -387MB (now 267MB) reserved=2560MB peak=1598MB [11:26] [ComfyUI] [SubprocessWorker] got response: ok [11:26] [ComfyUI] [comfy-env] DepthAnythingV3_MultiView done (3.77s): result: NodeOutput [11:26] [ComfyUI] [VRAM] After DepthAnythingV3_MultiView: 3207 / 24124 MB [11:26] [ComfyUI] [VRAM] Loaded: SubprocessModel (258 MB) [11:26] [ComfyUI] Prompt executed in 22.84 seconds [11:27] Node executed (3 total), capturing... [11:27] Frame 9 saved (frame_008.jpg, t=23.3s) [11:27] [capture-node] freeze=4ms shot=207ms unfreeze=7ms saved=True [11:27] Execution complete (t=23.7s) [11:27] Captured 9 unique frames over 23.76s [11:27] [debug] iframes: 0 [11:27] [timing] trigger_3d_previews: 0.0s [11:30] Saved high-quality screenshot: video_multiview_depth_executed.png [11:30] [timing] final_screenshot: 0.4s [11:30] [ComfyUI] [comfy-env] [VRAM] Before offload 'DepthAnything3Net_1' (258 MB) -> cpu: 3207 / 24124 MB [11:30] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ping (timeout=5.0s)... [11:30] [ComfyUI] [ComfyUI-DepthAnythingV3] Health check: ok [11:30] [ComfyUI] [SubprocessWorker] call_id=41 sending model_to_device [11:30] Results saved to /logs/DepthAnythingV3-0759/dev/linux-gpu/results.json [11:30] Model report: 9 files, 2.4 GB -> /logs/DepthAnythingV3-0759/dev/linux-gpu/models.json [11:30] Saved: /logs/DepthAnythingV3-0759/dev/linux-gpu/index.html [11:30] [EXECUTION] PASSED [11:30] linux: PASSED [11:30] Stopping ComfyUI server...