** ComfyUI Base Folder Path: D:\ComfyUI\ComfyUI ** User directory: D:\ComfyUI\ComfyUI\user ** ComfyUI-Manager config path: D:\ComfyUI\ComfyUI\user\__manager\config.ini ** Log path: D:\ComfyUI\ComfyUI\user\comfyui.log [INFO] [PRE] ComfyUI-Manager [INFO] Prestartup times for custom nodes: [INFO] 0.0 seconds: D:\ComfyUI\ComfyUI\custom_nodes\ComfyUI-MiniMax-H3-Studio-main [INFO] [INFO] Found comfy_kitchen backend eager: {'available': True, 'disabled': False, 'unavailable_reason': None, 'capabilities': ['adaln', 'apply_rope', 'apply_rope1', 'apply_rope1_', 'apply_rope_', 'apply_rope_split_half', 'apply_rope_split_half1', 'apply_rope_split_half1_', 'apply_rope_split_half_', 'convrot_w4a4_linear', 'dequantize_convrot_w4a4_weight', 'dequantize_int8_convrot_weight', 'dequantize_int8_convrot_weight_dtype', 'dequantize_int8_embedding', 'dequantize_int8_simple', 'dequantize_int8_simple_dtype', 'dequantize_mxfp8', 'dequantize_nvfp4', 'dequantize_per_tensor_fp8', 'dequantize_w4a8_int8_weight', 'fp16_conv3d', 'fp16_conv3d_out', 'fp16_linear', 'gemv_awq_w4a16', 'group_norm_silu_pad3d', 'group_norm_silu_pad3d_out', 'group_norm_stats3d', 'int8_linear', 'int8_linear_modulated', 'int8_linear_rms_modulated', 'na3d', 'prepare_int4_weight_for_int8_linear', 'quantize_and_rotate_rowwise', 'quantize_convrot_w4a4_weight', 'quantize_int8_convrot_weight', 'quantize_int8_rowwise', 'quantize_int8_tensorwise', 'quantize_mxfp8', 'quantize_nvfp4', 'quantize_per_tensor_fp8', 'quantize_svdquant_w4a4', 'quantize_w4a8_int8_weight', 'rms_adaln', 'rms_gated_residual', 'rms_rope', 'rms_rope1', 'rms_rope1_', 'rms_rope_', 'rms_rope_split_half', 'rms_rope_split_half1', 'rms_rope_split_half1_', 'rms_rope_split_half_', 'rotate_int8_convrot_weight', 'scaled_mm_mxfp8', 'scaled_mm_nvfp4', 'scaled_mm_svdquant_w4a4', 'sol_attn', 'stochastic_rounding_fp8', 'w4a8_int8_linear', 'w4a8_packed_linear', 'window_attention']} [INFO] Found comfy_kitchen backend triton: {'available': True, 'disabled': True, 'unavailable_reason': None, 'capabilities': ['adaln', 'apply_rope', 'apply_rope1', 'apply_rope1_', 'apply_rope_', 'apply_rope_split_half', 'apply_rope_split_half1', 'apply_rope_split_half1_', 'apply_rope_split_half_', 'dequantize_nvfp4', 'dequantize_per_tensor_fp8', 'int8_linear', 'na3d', 'quantize_and_rotate_rowwise', 'quantize_int8_rowwise', 'quantize_mxfp8', 'quantize_nvfp4', 'quantize_per_tensor_fp8', 'rms_adaln', 'rms_rope', 'rms_rope1', 'rms_rope1_', 'rms_rope_', 'rms_rope_split_half', 'rms_rope_split_half1', 'rms_rope_split_half1_', 'rms_rope_split_half_', 'w4a8_int8_linear']} [INFO] Found comfy_kitchen backend cuda: {'available': True, 'disabled': False, 'unavailable_reason': None, 'capabilities': ['adaln', 'apply_rope', 'apply_rope1', 'apply_rope1_', 'apply_rope_', 'apply_rope_split_half', 'apply_rope_split_half1', 'apply_rope_split_half1_', 'apply_rope_split_half_', 'closest_hit_on_mesh', 'closest_point_on_mesh', 'connected_components', 'convrot_w4a4_linear', 'delaunay3d', 'dequantize_convrot_w4a4_weight', 'dequantize_int8_convrot_weight', 'dequantize_int8_convrot_weight_dtype', 'dequantize_int8_simple', 'dequantize_int8_simple_dtype', 'dequantize_nvfp4', 'dequantize_per_tensor_fp8', 'dequantize_w4a8_int8_weight', 'edge_collapse_checks', 'fp16_conv3d', 'fp16_conv3d_out', 'fp16_linear', 'gemv_awq_w4a16', 'group_norm_silu_pad3d', 'group_norm_silu_pad3d_out', 'group_norm_stats3d', 'int8_linear', 'mesh_bvh', 'min_cut', 'na3d', 'prepare_int4_weight_for_int8_linear', 'quantize_and_rotate_rowwise', 'quantize_convrot_w4a4_weight', 'quantize_int8_convrot_weight', 'quantize_int8_rowwise', 'quantize_int8_tensorwise', 'quantize_mxfp8', 'quantize_nvfp4', 'quantize_per_tensor_fp8', 'quantize_svdquant_w4a4', 'quantize_w4a8_int8_weight', 'rms_adaln', 'rms_rope', 'rms_rope1', 'rms_rope1_', 'rms_rope_', 'rms_rope_split_half', 'rms_rope_split_half1', 'rms_rope_split_half1_', 'rms_rope_split_half_', 'rotate_int8_convrot_weight', 'scaled_mm_nvfp4', 'scaled_mm_svdquant_w4a4', 'sol_attn', 'stochastic_rounding_fp8', 'w4a8_int8_linear', 'w4a8_packed_linear', 'window_attention']} [INFO] Found comfy_kitchen backend ascend: {'available': False, 'disabled': False, 'unavailable_reason': "torch-npu is not installed: No module named 'torch_npu'", 'capabilities': []} [INFO] Found comfy_kitchen backend hip: {'available': False, 'disabled': False, 'unavailable_reason': 'PyTorch ROCm/HIP runtime not available', 'capabilities': []} [INFO] Checkpoint files will always be loaded safely. [INFO] Total VRAM 12288 MB, total RAM 130743 MB [INFO] pytorch version: 2.12.1+cu130 [INFO] Set vram state to: NORMAL_VRAM [INFO] Device: cuda:0 NVIDIA GeForce RTX 3060 : cudaMallocAsync [INFO] Device: cuda:1 NVIDIA GeForce RTX 4060 : cudaMallocAsync [INFO] Device: cuda:2 NVIDIA GeForce RTX 5060 Ti : cudaMallocAsync [INFO] Using async weight offloading with 2 streams [INFO] Enabled pinned memory 52297.0 [INFO] Using pytorch attention [INFO] Using Comfy Kitchen attention [INFO] aimdo: src-win/cuda-detour.c:38:INFO:aimdo_setup_hooks: installing 6 hooks [INFO] aimdo: src/control.c:269:INFO:comfy-aimdo NVML pressure enabled [INFO] aimdo: src-win/shmem-detect.c:76:INFO:comfy-aimdo DXCore adapter match: NVIDIA GeForce RTX 3060 runtime_luid=D00000000:00018f04 [INFO] aimdo: src-win/shmem-detect.c:112:INFO:comfy-aimdo WDDM adapter match: NVIDIA GeForce RTX 3060 runtime_luid=00000000:00018f04 dxgi_luid=00000000:00018f04 [INFO] aimdo: src/control.c:284:INFO:comfy-aimdo inited for GPU: NVIDIA GeForce RTX 3060 (VRAM: 12287 MB) [INFO] aimdo: src/control.c:269:INFO:comfy-aimdo NVML pressure enabled [INFO] aimdo: src-win/shmem-detect.c:76:INFO:comfy-aimdo DXCore adapter match: NVIDIA GeForce RTX 4060 runtime_luid=00000000:0001b56f [INFO] aimdo: src-win/shmem-detect.c:112:INFO:comfy-aimdo WDDM adapter match: NVIDIA GeForce RTX 4060 runtime_luid=00000000:0001b56f dxgi_luid=00000000:0001b56f [INFO] aimdo: src/control.c:284:INFO:comfy-aimdo inited for GPU: NVIDIA GeForce RTX 4060 (VRAM: 8187 MB) [INFO] aimdo: src/control.c:269:INFO:comfy-aimdo NVML pressure enabled [INFO] aimdo: src-win/shmem-detect.c:76:INFO:comfy-aimdo DXCore adapter match: NVIDIA GeForce RTX 5060 Ti runtime_luid=00000000:00015e73 [INFO] aimdo: src-win/shmem-detect.c:112:INFO:comfy-aimdo WDDM adapter match: NVIDIA GeForce RTX 5060 Ti runtime_luid=00000000:00015e73 dxgi_luid=00000000:00015e73 [INFO] aimdo: src/control.c:284:INFO:comfy-aimdo inited for GPU: NVIDIA GeForce RTX 5060 Ti (VRAM: 16283 MB) [INFO] DynamicVRAM support detected and enabled [INFO] Python version: 3.13.12 (main, Feb 12 2026, 00:38:53) [MSC v.1944 64 bit (AMD64)] [INFO] ComfyUI version: 0.45.4 [INFO] comfy-aimdo version: 0.6.0 [INFO] comfy-kitchen version: 0.2.41 [INFO] comfyui-frontend-package version: 1.59.0.dev2 [INFO] comfyui-workflow-templates version: 0.11.109 [INFO] comfyui-embedded-docs version: 0.5.16 [INFO] comfy-aimdo version: 0.6.0 [INFO] comfy-kitchen version: 0.2.41 [INFO] [Prompt Server] web root: D:\ComfyUI\ComfyUI\.venv\Lib\site-packages\comfyui_frontend_package\static [INFO] [assets-event] assets.enabled hashing_enabled=false [INFO] [START] ComfyUI-Manager [ComfyUI-Manager] Using GitPython backend [INFO] [ComfyUI-Manager] network_mode: public [WARNING] [ComfyUI-Manager] The matrix sharing feature has been disabled because the `matrix-nio` dependency is not installed. To use this feature, please run the following command: D:\ComfyUI\ComfyUI\.venv\Scripts\python.exe -m pip install matrix-nio [INFO] No OpenGL_accelerate module loaded: No module named 'OpenGL_accelerate' [INFO] ComfyUI-GGUF: Allowing full torch compile [AutoMultiGPU] ComfyUI-GGUF integration activated for AutoMultiGPU! [INFO] ComfyUI-GGUF integration activated for AutoMultiGPU! [AutoMultiGPU] ================================================== [INFO] ================================================== [AutoMultiGPU] [AutoMultiGPU] ComfyUI-AutoMultiGPU loaded with 3 Production Engines: [INFO] [AutoMultiGPU] ComfyUI-AutoMultiGPU loaded with 3 Production Engines: [AutoMultiGPU] 1. AutoMultiGPU - DisTorch Engine (Layer Sharding) [INFO] 1. AutoMultiGPU - DisTorch Engine (Layer Sharding) [AutoMultiGPU] 2. AutoMultiGPU - PyTorch Engine (Native Device Split) [INFO] 2. AutoMultiGPU - PyTorch Engine (Native Device Split) [AutoMultiGPU] 3. AutoMultiGPU - Automatic Engine (Zero-OOM Governor) [INFO] 3. AutoMultiGPU - Automatic Engine (Zero-OOM Governor) [AutoMultiGPU] Detected cuda:0: NVIDIA GeForce RTX 3060 | Total: 12.0 GB | Free: 10.99 GB [INFO] Detected cuda:0: NVIDIA GeForce RTX 3060 | Total: 12.0 GB | Free: 10.99 GB [AutoMultiGPU] Detected cuda:1: NVIDIA GeForce RTX 4060 | Total: 8.0 GB | Free: 6.94 GB [INFO] Detected cuda:1: NVIDIA GeForce RTX 4060 | Total: 8.0 GB | Free: 6.94 GB [AutoMultiGPU] Detected cuda:2: NVIDIA GeForce RTX 5060 Ti | Total: 15.9 GB | Free: 14.82 GB [INFO] Detected cuda:2: NVIDIA GeForce RTX 5060 Ti | Total: 15.9 GB | Free: 14.82 GB [AutoMultiGPU] ================================================== [INFO] ================================================== [INFO] Qwen2D VAE patch is disabled by default (v2.8.3). Set DYPE_ENABLE_QWEN2D_VAE=1 to enable it. [INFO] ComfyUI-GGUF: Allowing full torch compile [INFO] ### Loading: ComfyUI-Impact-Pack (V8.30) [INFO] [Impact Pack] Wildcard total size (0.00 MB) is within cache limit (50.00 MB). Using full cache mode. [INFO] [Impact Pack] Wildcards loading done. [INFO] [H3 Studio] Face Refine sampler: source-latent FL2VA bridge enabled [INFO] [H3 Studio] Face Refine integration: final-still stage enabled [INFO] [H3 Studio] Consolidated integrity v18 installed [INFO] [H3 Studio] Post-merge v21 conditioning/role/preview fixes installed [INFO] [MultiGPU Core Patching] Patching mm.soft_empty_cache for Comprehensive Memory Management (VRAM + CPU + Store Pruning) [MultiGPU Core Patching] Patching mm.get_torch_device, mm.text_encoder_device, mm.unet_offload_device [MultiGPU DEBUG] Initial current_device: cuda:0 [MultiGPU DEBUG] Initial current_text_encoder_device: cuda:0 [MultiGPU DEBUG] Initial current_unet_offload_device: cpu [MultiGPU] Patched comfy.model_management.current_stream to honor CUDA device arguments [MultiGPU] Patched comfy.sample.sample with runtime device guard [MultiGPU] Patched comfy.sample.sample_custom with runtime device guard [MultiGPU] comfy_aimdo already initialized for CUDA devices [0, 1, 2] [MultiGPU] Initiating custom_node Registration. . . ----------------------------------------------- custom_node Found Nodes ----------------------------------------------- ComfyUI-LTXVideo N 0 ComfyUI-Florence2 N 0 ComfyUI_bitsandbytes_NF4 N 0 x-flux-comfyui N 0 ComfyUI-MMAudio N 0 ComfyUI-GGUF Y 18 PuLID_ComfyUI N 0 ComfyUI-WanVideoWrapper N 0 ----------------------------------------------- [MultiGPU] Registration complete. Final mappings: CheckpointLoaderAdvancedMultiGPU, CheckpointLoaderAdvancedDisTorch2MultiGPU, DeviceSelectorMultiGPU, UNetLoaderLP, UNETLoaderMultiGPU, VAELoaderMultiGPU, CLIPLoaderMultiGPU, DualCLIPLoaderMultiGPU, TripleCLIPLoaderMultiGPU, QuadrupleCLIPLoaderMultiGPU, CLIPVisionLoaderMultiGPU, CheckpointLoaderSimpleMultiGPU, ControlNetLoaderMultiGPU, DiffusersLoaderMultiGPU, DiffControlNetLoaderMultiGPU, UNETLoaderDisTorch2MultiGPU, VAELoaderDisTorch2MultiGPU, CLIPLoaderDisTorch2MultiGPU, DualCLIPLoaderDisTorch2MultiGPU, TripleCLIPLoaderDisTorch2MultiGPU, QuadrupleCLIPLoaderDisTorch2MultiGPU, CLIPVisionLoaderDisTorch2MultiGPU, CheckpointLoaderSimpleDisTorch2MultiGPU, ControlNetLoaderDisTorch2MultiGPU, DiffusersLoaderDisTorch2MultiGPU, DiffControlNetLoaderDisTorch2MultiGPU, UnetLoaderGGUFDisTorchMultiGPU, UnetLoaderGGUFAdvancedDisTorchMultiGPU, CLIPLoaderGGUFDisTorchMultiGPU, DualCLIPLoaderGGUFDisTorchMultiGPU, TripleCLIPLoaderGGUFDisTorchMultiGPU, QuadrupleCLIPLoaderGGUFDisTorchMultiGPU, UnetLoaderGGUFDisTorch2MultiGPU, UnetLoaderGGUFAdvancedDisTorch2MultiGPU, CLIPLoaderGGUFDisTorch2MultiGPU, DualCLIPLoaderGGUFDisTorch2MultiGPU, TripleCLIPLoaderGGUFDisTorch2MultiGPU, QuadrupleCLIPLoaderGGUFDisTorch2MultiGPU, UnetLoaderGGUFMultiGPU, UnetLoaderGGUFAdvancedMultiGPU, CLIPLoaderGGUFMultiGPU, DualCLIPLoaderGGUFMultiGPU, TripleCLIPLoaderGGUFMultiGPU, QuadrupleCLIPLoaderGGUFMultiGPU [INFO] MiniMax H3 Director HTTP routes registered [INFO] Import times for custom nodes: [INFO] 0.0 seconds: D:\ComfyUI\ComfyUI\custom_nodes\websocket_image_save.py [INFO] 0.0 seconds: D:\ComfyUI\ComfyUI\custom_nodes\comfyui-SelfLift-master [INFO] 0.0 seconds: D:\ComfyUI\ComfyUI\custom_nodes\ComfyUI-GGUF [INFO] 0.0 seconds: D:\ComfyUI\ComfyUI\custom_nodes\ComfyUI-H3-FaceRefine [INFO] 0.0 seconds: D:\ComfyUI\ComfyUI\custom_nodes\Comfyui-MMH3-UltimateUpscale-main [INFO] 0.0 seconds: D:\ComfyUI\ComfyUI\custom_nodes\Comfyui-MMH3-UltimateUpscale [INFO] 0.0 seconds: D:\ComfyUI\ComfyUI\custom_nodes\comfyui-minimax-h3-latent-upscaler [INFO] 0.0 seconds: D:\ComfyUI\ComfyUI\custom_nodes\ComfyUI-MultiGPU-main [INFO] 0.0 seconds: D:\ComfyUI\ComfyUI\custom_nodes\ComfyUI_MiniMaxH3_Director [INFO] 0.0 seconds: D:\ComfyUI\ComfyUI\custom_nodes\comfyui-kjnodes [INFO] 0.0 seconds: D:\ComfyUI\ComfyUI\custom_nodes\ComfyUI-DyPE-main [INFO] 0.0 seconds: D:\ComfyUI\ComfyUI\custom_nodes\ComfyUI-H3-NativeAudioLock_ND [INFO] 0.0 seconds: D:\ComfyUI\ComfyUI\custom_nodes\ComfyUI-MiniMax-H3-Studio-main [INFO] 0.1 seconds: D:\ComfyUI\ComfyUI\custom_nodes\ComfyUI-MiniMaxH3-Director [INFO] 0.1 seconds: D:\ComfyUI\ComfyUI\custom_nodes\comfyui-videohelpersuite [INFO] 0.1 seconds: D:\ComfyUI\ComfyUI\custom_nodes\comfyui-impact-pack [INFO] 0.1 seconds: D:\ComfyUI\ComfyUI\custom_nodes\comfyui-automultigpu [INFO] 0.8 seconds: D:\ComfyUI\ComfyUI\custom_nodes\ComfyUI_Nvidia_VFX_Nodes [INFO] 1.0 seconds: D:\ComfyUI\ComfyUI\custom_nodes\minimax-h3-audio-T8 [INFO] [INFO] Context impl SQLiteImpl. [INFO] Will assume non-transactional DDL. [INFO] Seeder start (roots=('models', 'input', 'output'), phase=full) [INFO] [assets-event] seeder.scan_started phase=full [INFO] Background asset scan initiated for models, input, output [INFO] [assets-event] seeder.marked_missing count=0 stage=pruning [INFO] Asset scan [models] directory: D:\ComfyUI\ComfyUI\models [INFO] Asset scan [input] directories: ['D:\\Comfy-Desktop\\ComfyUI-Shared\\input'] [INFO] Asset scan [output] directories: ['D:\\Comfy-Desktop\\ComfyUI-Shared\\output'] [INFO] Using RAM pressure cache. [INFO] Starting server [INFO] To see the GUI go to: http://127.0.0.1:8188 [INFO] Fast scan complete: 0.151s total (created=1, skipped=175, total_paths=177) [INFO] Scan(('models', 'input', 'output'), full) done 0.196s: created=1 enriched=1 skipped=175 [INFO] [assets-event] seeder.scan_completed cpu_ms=62 created=1 dirs_listed_count=13 elapsed_ms=196 enrich_failed=0 enriched=1 files_statted_count=177 hash_failed=0 missing_marked_count=0 paused_ms=0 permission_denied=0 phase=full recovered_count=0 skipped=175 [INFO] Seeder start (roots=('models', 'input', 'output'), phase=full) [INFO] [assets-event] seeder.scan_started phase=full [INFO] Asset scan [models] directory: D:\ComfyUI\ComfyUI\models [INFO] Asset scan [input] directories: ['D:\\Comfy-Desktop\\ComfyUI-Shared\\input'] [INFO] Asset scan [output] directories: ['D:\\Comfy-Desktop\\ComfyUI-Shared\\output'] [INFO] Fast scan complete: 0.039s total (created=0, skipped=176, total_paths=177) [INFO] Scan(('models', 'input', 'output'), full) done 0.054s: created=0 enriched=0 skipped=176 [INFO] [assets-event] seeder.scan_completed cpu_ms=47 created=0 dirs_listed_count=13 elapsed_ms=54 enrich_failed=0 enriched=0 files_statted_count=174 hash_failed=0 missing_marked_count=0 paused_ms=0 permission_denied=0 phase=full recovered_count=0 skipped=176 [MMH3-UltimateUpscale] Ignored 1 pickle checkpoint(s) in the latent_upscale_models folder (only .safetensors is supported): minimax_h3_latent_upscaler_3d_conv_v1_fp32.pth [INFO] Seeder start (roots=('models',), phase=full) [INFO] [assets-event] seeder.scan_started phase=full root=models [INFO] Asset scan [models] directory: D:\ComfyUI\ComfyUI\models [INFO] Fast scan complete: 0.041s total (created=0, skipped=47, total_paths=48) [INFO] Scan(('models',), full) done 0.058s: created=0 enriched=0 skipped=47 [INFO] [assets-event] seeder.scan_completed cpu_ms=47 created=0 dirs_listed_count=0 elapsed_ms=58 enrich_failed=0 enriched=0 files_statted_count=45 hash_failed=0 missing_marked_count=0 paused_ms=0 permission_denied=0 phase=full recovered_count=0 root=models skipped=47 [INFO] The local agent runs comfy-agent-go 0.1.0-rc9 (D:\ComfyUI\ComfyUI\.venv\Lib\site-packages\comfy_agent_go\_bin\comfy-agent.exe). [INFO] [agent-event] agent_starting agent_version=0.1.0-rc9 attempt=1 python_arch=amd64 python_version=3.13.12 [INFO] [comfy-agent] {"level":"info","endpoint":"https://cloud.comfy.org/api/agent/llm/v1/traces","content":false,"time":"2026-11-16T12:39:23+08:00","message":"telemetry: exporting traces"} [INFO] [comfy-agent] {"level":"info","step":"open data dir","step_id":"open_data_dir","expected":0,"time":"2026-11-16T12:39:23+08:00","message":"startup step started"} [INFO] [comfy-agent] {"level":"info","step":"open data dir","outcome":"ok","duration_ms":6,"time":"2026-11-16T12:39:23+08:00","message":"startup step finished"} [INFO] [comfy-agent] {"level":"info","step":"find ComfyUI","step_id":"find_comfyui","expected":0,"time":"2026-11-16T12:39:23+08:00","message":"startup step started"} [INFO] [agent-event] startup_step step=open_data_dir outcome=ok duration_ms=6 [INFO] [comfy-agent] {"level":"info","step":"find ComfyUI","outcome":"ok","duration_ms":12,"time":"2026-11-16T12:39:23+08:00","message":"startup step finished"} [INFO] [agent-event] startup_step step=find_comfyui outcome=ok duration_ms=12 [INFO] [comfy-agent] {"level":"info","path":"D:\\ComfyUI\\ComfyUI","node_trees":["D:\\ComfyUI\\ComfyUI\\custom_nodes"],"found_by":"AGENT_COMFY_PATH","comfy_url":"http://127.0.0.1:8188","time":"2026-11-16T12:39:23+08:00","message":"ComfyUI install"} [INFO] [comfy-agent] {"level":"info","step":"find Node","step_id":"find_node","expected":360000,"time":"2026-11-16T12:39:23+08:00","message":"startup step started"} [INFO] [comfy-agent] {"level":"info","node":"D:\\ComfyUI\\ComfyUI\\.venv\\Lib\\site-packages\\comfy_agent_node\\_bin\\node.exe","node_version":"v22.23.2","time":"2026-11-16T12:39:23+08:00","message":"Node for the document host"} [INFO] [agent-event] node_found node_version=v22.23.2 [INFO] [comfy-agent] {"level":"info","step":"find Node","outcome":"ok","duration_ms":22,"time":"2026-11-16T12:39:23+08:00","message":"startup step finished"} [INFO] [comfy-agent] {"level":"info","step":"sandbox config","step_id":"sandbox_config","expected":0,"time":"2026-11-16T12:39:23+08:00","message":"startup step started"} [INFO] [agent-event] startup_step step=find_node outcome=ok duration_ms=22 [INFO] [comfy-agent] {"level":"info","step":"sandbox config","outcome":"ok","duration_ms":56,"time":"2026-11-16T12:39:23+08:00","message":"startup step finished"} [INFO] [agent-event] startup_step step=sandbox_config outcome=ok duration_ms=56 [INFO] [comfy-agent] {"level":"info","step":"egress proxy","step_id":"egress_proxy","expected":0,"time":"2026-11-16T12:39:23+08:00","message":"startup step started"} [INFO] [comfy-agent] {"level":"info","step":"egress proxy","outcome":"ok","duration_ms":0,"time":"2026-11-16T12:39:23+08:00","message":"startup step finished"} [INFO] [comfy-agent] {"level":"info","step":"sandbox grants","step_id":"sandbox_grants","expected":0,"time":"2026-11-16T12:39:23+08:00","message":"startup step started"} [INFO] [agent-event] startup_step step=egress_proxy outcome=ok duration_ms=0 [INFO] [comfy-agent] {"level":"info","granted":6,"wanted":6,"duration_ms":0,"time":"2026-11-16T12:39:23+08:00","message":"windows appcontainer: sandbox grants applied"} [INFO] [comfy-agent] {"level":"info","step":"sandbox grants","outcome":"ok","duration_ms":100,"time":"2026-11-16T12:39:24+08:00","message":"startup step finished"} [INFO] [agent-event] startup_step step=sandbox_grants outcome=ok duration_ms=100 [INFO] [comfy-agent] {"level":"info","comfy_url":"http://127.0.0.1:8188","assets":"enabled","time":"2026-11-16T12:39:24+08:00","message":"ComfyUI asset system"} [INFO] [comfy-agent] {"level":"info","step":"read node catalog from ComfyUI","step_id":"object_info","expected":120000,"time":"2026-11-16T12:39:24+08:00","message":"startup step started"} [INFO] Seeder start (roots=('models', 'input', 'output'), phase=full) [INFO] [assets-event] seeder.scan_started phase=full [INFO] Asset scan [models] directory: D:\ComfyUI\ComfyUI\models [INFO] Asset scan [input] directories: ['D:\\Comfy-Desktop\\ComfyUI-Shared\\input'] [INFO] Asset scan [output] directories: ['D:\\Comfy-Desktop\\ComfyUI-Shared\\output'] [INFO] Fast scan complete: 0.349s total (created=0, skipped=176, total_paths=177) [INFO] Scan(('models', 'input', 'output'), full) done 0.369s: created=0 enriched=0 skipped=176 [INFO] [assets-event] seeder.scan_completed cpu_ms=359 created=0 dirs_listed_count=13 elapsed_ms=369 enrich_failed=0 enriched=0 files_statted_count=174 hash_failed=0 missing_marked_count=0 paused_ms=0 permission_denied=0 phase=full recovered_count=0 skipped=176 [MMH3-UltimateUpscale] Ignored 1 pickle checkpoint(s) in the latent_upscale_models folder (only .safetensors is supported): minimax_h3_latent_upscaler_3d_conv_v1_fp32.pth [INFO] [comfy-agent] {"level":"info","step":"read node catalog from ComfyUI","outcome":"ok","duration_ms":596,"time":"2026-11-16T12:39:24+08:00","message":"startup step finished"} [INFO] [comfy-agent] {"level":"info","step":"load knowledge","step_id":"knowledge","expected":0,"time":"2026-11-16T12:39:24+08:00","message":"startup step started"} [INFO] [agent-event] startup_step step=object_info outcome=ok duration_ms=596 [WARNING] [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. [INFO] Seeder start (roots=('models', 'input', 'output'), phase=full) [INFO] [assets-event] seeder.scan_started phase=full [INFO] Asset scan [models] directory: D:\ComfyUI\ComfyUI\models [INFO] Asset scan [input] directories: ['D:\\Comfy-Desktop\\ComfyUI-Shared\\input'] [INFO] Asset scan [output] directories: ['D:\\Comfy-Desktop\\ComfyUI-Shared\\output'] [INFO] Fast scan complete: 0.038s total (created=0, skipped=176, total_paths=177) [INFO] Scan(('models', 'input', 'output'), full) done 0.090s: created=0 enriched=0 skipped=176 [INFO] [assets-event] seeder.scan_completed cpu_ms=47 created=0 dirs_listed_count=13 elapsed_ms=90 enrich_failed=0 enriched=0 files_statted_count=174 hash_failed=0 missing_marked_count=0 paused_ms=0 permission_denied=0 phase=full recovered_count=0 skipped=176 [MMH3-UltimateUpscale] Ignored 1 pickle checkpoint(s) in the latent_upscale_models folder (only .safetensors is supported): minimax_h3_latent_upscaler_3d_conv_v1_fp32.pth [INFO] [comfy-agent] {"level":"info","step":"load knowledge","outcome":"ok","duration_ms":680,"time":"2026-11-16T12:39:25+08:00","message":"startup step finished"} [INFO] [agent-event] startup_step step=knowledge outcome=ok duration_ms=680 [INFO] [comfy-agent] {"level":"info","step":"import comfy CLI skills","step_id":"cli_skills","expected":0,"time":"2026-11-16T12:39:25+08:00","message":"startup step started"} [INFO] [ComfyUI-Manager] The ComfyRegistry cache update is still in progress, so an outdated cache is being used. FETCH DATA from: D:\ComfyUI\ComfyUI\.venv\Lib\site-packages\comfyui_manager\custom-node-list.json [DONE] [ERROR] Exception in callback _ProactorBasePipeTransport._call_connection_lost() handle: Traceback (most recent call last): File "D:\ComfyUI\standalone-env\Lib\asyncio\events.py", line 89, in _run self._context.run(self._callback, *self._args) ~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\standalone-env\Lib\asyncio\proactor_events.py", line 165, in _call_connection_lost self._sock.shutdown(socket.SHUT_RDWR) ~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^ ConnectionResetError: [WinError 10054] 遠端主機已強制關閉一個現存的連線。 [WARNING] [comfy-agent] {"level":"warn","max_packs":16,"reason":"too_many_packs","time":"2026-11-16T12:39:29+08:00","message":"comfy CLI skill import: pack limit reached; skipping the rest"} [INFO] [comfy-agent] {"level":"info","imported":16,"skipped":7,"listed":23,"time":"2026-11-16T12:39:29+08:00","message":"comfy CLI skill import complete"} [INFO] [comfy-agent] {"level":"info","step":"import comfy CLI skills","outcome":"ok","duration_ms":4578,"time":"2026-11-16T12:39:29+08:00","message":"startup step finished"} [WARNING] [comfy-agent] {"level":"warn","skill":"comfy-director","time":"2026-11-16T12:39:29+08:00","message":"comfy CLI skill import: this name is already a first-party pack; keeping the first-party one"} [INFO] [agent-event] startup_step step=cli_skills outcome=ok duration_ms=4578 [INFO] [comfy-agent] {"level":"info","step":"detect environment","step_id":"detect_environment","expected":0,"time":"2026-11-16T12:39:29+08:00","message":"startup step started"} [INFO] [comfy-agent] {"level":"info","step":"detect environment","outcome":"ok","duration_ms":837,"time":"2026-11-16T12:39:30+08:00","message":"startup step finished"} [INFO] [agent-event] startup_step step=detect_environment outcome=ok duration_ms=837 [INFO] [comfy-agent] {"level":"info","step":"start document host","step_id":"document_host","expected":0,"time":"2026-11-16T12:39:30+08:00","message":"startup step started"} [INFO] [comfy-agent] {"level":"debug","source":"doc-host","time":"2026-11-16T12:39:30+08:00","message":"doc-host listening on 127.0.0.1:59330 (schema_version=5)"} [INFO] [comfy-agent] {"level":"info","endpoint":"http://127.0.0.1:59330","pid":14156,"node":"D:\\ComfyUI\\ComfyUI\\.venv\\Lib\\site-packages\\comfy_agent_node\\_bin\\node.exe","node_version":"v22.23.2","time":"2026-11-16T12:39:31+08:00","message":"doc host launched on loopback (local CRDT write path available)"} [INFO] [comfy-agent] {"level":"info","step":"start document host","outcome":"ok","duration_ms":679,"time":"2026-11-16T12:39:31+08:00","message":"startup step finished"} [INFO] [agent-event] startup_step step=document_host outcome=ok duration_ms=679 [INFO] [comfy-agent] {"level":"info","step":"listen","step_id":"listen","expected":0,"time":"2026-11-16T12:39:31+08:00","message":"startup step started"} [INFO] [comfy-agent] {"level":"info","step":"listen","outcome":"ok","duration_ms":1,"time":"2026-11-16T12:39:31+08:00","message":"startup step finished"} [INFO] [comfy-agent] {"level":"info","port":59335,"health":"http://127.0.0.1:59335/health","data_dir":"D:\\ComfyUI\\ComfyUI\\user\\__agent","version":"0.1.0-rc9","os":"windows","os_build":"10.0.26300 amd64","crdt":"on","stale_schema_reseed":"on","stale_schema_upgrade":"on","time":"2026-11-16T12:39:31+08:00","message":"comfy-agent ready"} [INFO] [agent-event] startup_step step=listen outcome=ok duration_ms=1 [INFO] [comfy-agent] ⇨ http server started on 127.0.0.1:59335 [INFO] [comfy-agent] {"level":"info","remote_ip":"127.0.0.1","host":"127.0.0.1:59335","method":"GET","uri":"/health","request_id":"","user_agent":"Python-urllib/3.13","status":200,"content_length":"","response_size":1790,"latency":0,"time":"2026-11-16T12:39:31+08:00","message":"request"} [INFO] [agent-event] agent_started duration_ms=8518 agent_version=0.1.0-rc9 startup_ms=7710 crdt=on cli=override comfyui=found comfyui_found_by=explicit comfyui_match=match shell=on sandbox=appcontainer assets=enabled [INFO] Local agent ready on 127.0.0.1:59335 (pid 28028). [INFO] FETCH ComfyRegistry Data [DONE] [INFO] [ComfyUI-Manager] default cache updated: https://api.comfy.org/nodes FETCH DATA from: D:\ComfyUI\ComfyUI\user\__manager\cache\1514988643_custom-node-list.json [DONE] [INFO] [ComfyUI-Manager] All startup tasks have been completed. 2D [INFO] got prompt [MultiGPU CUDA Guard] Switching CUDA current device 0 -> 1 (NodeOverride.load_vae) [INFO] Model storage policy: fast_disk=False paths=['D:\\Comfy-Desktop\\ComfyUI-Shared\\models\\vae\\minimax_h3_audio_vae_fp32.safetensors'] [INFO] VAE load device: cuda:1, offload device: cpu, dtype: torch.float32 [MultiGPU CUDA Guard] Restored CUDA current device 1 -> 0 (NodeOverride.load_vae) [MultiGPU Core Patching] Successfully patched ModelPatcher.partially_load [INFO] Model storage policy: fast_disk=False paths=['D:\\Comfy-Desktop\\ComfyUI-Shared\\models\\vae\\minimax_h3_video_vae_int8_convrot.safetensors'] [INFO] Found quantization metadata version 1 [INFO] VAE load device: cuda:0, offload device: cpu, dtype: torch.float16 [MultiGPU DisTorch V2] Full allocation string: #cuda:0;32.0;cpu [AutoMultiGPU] Loading CLIP [qwen3vl_32b_minimax_h3_nvfp4_awq.safetensors] on target device: cuda:0 (CPU Offload: auto_smart_offload, Model Type: minimax) [INFO] Loading CLIP [qwen3vl_32b_minimax_h3_nvfp4_awq.safetensors] on target device: cuda:0 (CPU Offload: auto_smart_offload, Model Type: minimax) [INFO] Found quantization metadata version 1 [MultiGPU Core Patching] text_encoder_device_patched returning device: cuda:0 (current_text_encoder_device=cuda:0) [INFO] Using MixedPrecisionOps for text encoder [INFO] Model storage policy: fast_disk=False paths=['D:\\Comfy-Desktop\\ComfyUI-Shared\\models\\text_encoders\\qwen3vl_32b_minimax_h3_nvfp4_awq.safetensors'] [INFO] CLIP/text encoder model load device: cuda:0, offload device: cpu, current: cpu, dtype: torch.float16 [INFO] Requested to load MiniMaxH3VideoVAE [INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 2677MB Staged. 0 patches attached. Force pre-loaded 128 weights: 691 KB. [INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 2677MB Staged. 0 patches attached. Force pre-loaded 128 weights: 691 KB. [INFO] Requested to load MiniMaxH3TEModel_ [INFO] Model MiniMaxH3TEModel_ prepared for dynamic VRAM loading. 14956MB Staged. 0 patches attached. Force pre-loaded 410 weights: 4572 KB. [AutoMultiGPU] Loading UNET [minimax_h3_ref2va_pruned_w4a8_mixed.safetensors] (Format: Safetensors) on target device: cuda:2 (CPU Offload: always_offload_to_cpu) [INFO] Loading UNET [minimax_h3_ref2va_pruned_w4a8_mixed.safetensors] (Format: Safetensors) on target device: cuda:2 (CPU Offload: always_offload_to_cpu) [INFO] Found quantization metadata version 1 [INFO] Detected mixed precision quantization [INFO] Using mixed precision operations [INFO] Native ops: w6a8_int8, convrot_w4a4, asym_w4a8_int8, int8_tensorwise , emulated ops: nvfp4, float8_e5m2, float8_e4m3fn, mxfp8 [INFO] model weight dtype torch.bfloat16, manual cast: torch.bfloat16 [INFO] model_type FLOW_AV [INFO] Model storage policy: fast_disk=False paths=['D:\\Comfy-Desktop\\ComfyUI-Shared\\models\\diffusion_models\\minimax_h3_ref2va_pruned_w4a8_mixed.safetensors'] [INFO] Using sage attention mode: auto [INFO] Requested to load MiniMaxH3 [INFO] 0 models unloaded. [INFO] Model MiniMaxH3 prepared for dynamic VRAM loading. 11221MB Staged. 208 patches attached. Force pre-loaded 210 weights: 1175 KB. 12%|█▎ | 1/8 [01:53<13:12, 113.18s/it, Model Initialization complete! ][INFO] FETCH ComfyRegistry Data [DONE] [INFO] [ComfyUI-Manager] default cache updated: https://api.comfy.org/nodes FETCH DATA from: D:\ComfyUI\ComfyUI\user\__manager\cache\1514988643_custom-node-list.json [DONE] [INFO] [ComfyUI-Manager] All startup tasks have been completed. 100%|██████████| 8/8 [15:41<00:00, 117.65s/it] [MinimaxH3-3D] Latent 80x46 -> 88x50 | Pixels 1408x800 | scale=1.100 [MinimaxH3-3D] Loaded upscale model: minimax_h3_latent_upscaler_3d_conv_v1_fp16.safetensors Params: 345,280,216 | Attn: forced off | Temporal: on (every=2, kernel=5) | Precision: fp16 | Device: cuda [INFO] Requested to load MiniMaxH3 [INFO] 0 models unloaded. [INFO] Model MiniMaxH3 prepared for dynamic VRAM loading. 11221MB Staged. 208 patches attached. Force pre-loaded 210 weights: 1175 KB. 100%|██████████| 4/4 [10:11<00:00, 152.81s/it] [INFO] Requested to load MiniMaxH3AudioVAE [INFO] Model MiniMaxH3AudioVAE prepared for dynamic VRAM loading. 576MB Staged. 0 patches attached. Force pre-loaded 401 weights: 539 KB. [INFO] Requested to load MiniMaxH3VideoVAE [INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 2677MB Staged. 0 patches attached. Force pre-loaded 128 weights: 691 KB. [INFO] Prompt executed in 00:28:07 [INFO] Seeder start (roots=('output',), phase=full) [INFO] [assets-event] seeder.scan_started phase=full root=output [INFO] Asset scan [output] directories: ['D:\\Comfy-Desktop\\ComfyUI-Shared\\output'] [INFO] Fast scan complete: 0.003s total (created=0, skipped=27, total_paths=27) [INFO] Scan(('output',), full) done 0.006s: created=0 enriched=0 skipped=27 [INFO] [assets-event] seeder.scan_completed cpu_ms=0 created=0 dirs_listed_count=10 elapsed_ms=6 enrich_failed=0 enriched=0 files_statted_count=0 hash_failed=0 missing_marked_count=0 paused_ms=0 permission_denied=0 phase=full recovered_count=0 root=output skipped=27 3D [INFO] got prompt [MultiGPU CUDA Guard] Switching CUDA current device 0 -> 1 (NodeOverride.load_vae) [INFO] Model storage policy: fast_disk=False paths=['D:\\Comfy-Desktop\\ComfyUI-Shared\\models\\vae\\minimax_h3_audio_vae_fp32.safetensors'] [INFO] VAE load device: cuda:1, offload device: cpu, dtype: torch.float32 [MultiGPU CUDA Guard] Restored CUDA current device 1 -> 0 (NodeOverride.load_vae) [MultiGPU Core Patching] Successfully patched ModelPatcher.partially_load [INFO] Model storage policy: fast_disk=False paths=['D:\\Comfy-Desktop\\ComfyUI-Shared\\models\\vae\\minimax_h3_video_vae_int8_convrot.safetensors'] [INFO] Found quantization metadata version 1 [INFO] VAE load device: cuda:0, offload device: cpu, dtype: torch.float16 [MultiGPU DisTorch V2] Full allocation string: #cuda:0;32.0;cpu [AutoMultiGPU] Loading CLIP [qwen3vl_32b_minimax_h3_nvfp4_awq.safetensors] on target device: cuda:0 (CPU Offload: auto_smart_offload, Model Type: minimax) [INFO] Loading CLIP [qwen3vl_32b_minimax_h3_nvfp4_awq.safetensors] on target device: cuda:0 (CPU Offload: auto_smart_offload, Model Type: minimax) [INFO] Found quantization metadata version 1 [MultiGPU Core Patching] text_encoder_device_patched returning device: cuda:0 (current_text_encoder_device=cuda:0) [INFO] Using MixedPrecisionOps for text encoder [INFO] Model storage policy: fast_disk=False paths=['D:\\Comfy-Desktop\\ComfyUI-Shared\\models\\text_encoders\\qwen3vl_32b_minimax_h3_nvfp4_awq.safetensors'] [INFO] CLIP/text encoder model load device: cuda:0, offload device: cpu, current: cpu, dtype: torch.float16 [INFO] Requested to load MiniMaxH3VideoVAE [INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 2677MB Staged. 0 patches attached. Force pre-loaded 128 weights: 691 KB. [INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 2677MB Staged. 0 patches attached. Force pre-loaded 128 weights: 691 KB. [INFO] Requested to load MiniMaxH3TEModel_ [INFO] Model MiniMaxH3TEModel_ prepared for dynamic VRAM loading. 14956MB Staged. 0 patches attached. Force pre-loaded 410 weights: 4572 KB. [AutoMultiGPU] Loading UNET [minimax_h3_ref2va_pruned_w4a8_mixed.safetensors] (Format: Safetensors) on target device: cuda:2 (CPU Offload: always_offload_to_cpu) [INFO] Loading UNET [minimax_h3_ref2va_pruned_w4a8_mixed.safetensors] (Format: Safetensors) on target device: cuda:2 (CPU Offload: always_offload_to_cpu) [INFO] Found quantization metadata version 1 [INFO] Detected mixed precision quantization [INFO] Using mixed precision operations [INFO] Native ops: int8_tensorwise, w6a8_int8, asym_w4a8_int8, convrot_w4a4 , emulated ops: nvfp4, float8_e5m2, mxfp8, float8_e4m3fn [INFO] model weight dtype torch.bfloat16, manual cast: torch.bfloat16 [INFO] model_type FLOW_AV [INFO] Model storage policy: fast_disk=False paths=['D:\\Comfy-Desktop\\ComfyUI-Shared\\models\\diffusion_models\\minimax_h3_ref2va_pruned_w4a8_mixed.safetensors'] [INFO] Using sage attention mode: auto [INFO] Requested to load MiniMaxH3 [INFO] 0 models unloaded. [INFO] Model MiniMaxH3 prepared for dynamic VRAM loading. 11221MB Staged. 208 patches attached. Force pre-loaded 210 weights: 1175 KB. 100%|██████████| 8/8 [15:45<00:00, 118.16s/it] [MinimaxH3-3D] Latent 80x46 -> 88x50 | Pixels 1408x800 | scale=1.100 [MinimaxH3-3D] Loaded upscale model: minimax_h3_latent_upscaler_3d_conv_v1_fp16.safetensors Params: 345,280,216 | Attn: forced off | Temporal: on (every=2, kernel=5) | Precision: fp16 | Device: cuda [INFO] Requested to load MiniMaxH3 [INFO] 0 models unloaded. [INFO] Model MiniMaxH3 prepared for dynamic VRAM loading. 11221MB Staged. 208 patches attached. Force pre-loaded 210 weights: 1175 KB. 100%|██████████| 4/4 [10:11<00:00, 152.90s/it] [INFO] Requested to load MiniMaxH3AudioVAE [INFO] Model MiniMaxH3AudioVAE prepared for dynamic VRAM loading. 576MB Staged. 0 patches attached. Force pre-loaded 401 weights: 539 KB. [INFO] Requested to load MiniMaxH3VideoVAE [INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 2677MB Staged. 0 patches attached. Force pre-loaded 128 weights: 691 KB. [INFO] Prompt executed in 00:28:13 [INFO] Seeder start (roots=('output',), phase=full) [INFO] [assets-event] seeder.scan_started phase=full root=output [INFO] Asset scan [output] directories: ['D:\\Comfy-Desktop\\ComfyUI-Shared\\output'] [INFO] Fast scan complete: 0.003s total (created=0, skipped=29, total_paths=29) [INFO] Scan(('output',), full) done 0.005s: created=0 enriched=0 skipped=29 [INFO] [assets-event] seeder.scan_completed cpu_ms=0 created=0 dirs_listed_count=10 elapsed_ms=5 enrich_failed=0 enriched=0 files_statted_count=0 hash_failed=0 missing_marked_count=0 paused_ms=0 permission_denied=0 phase=full recovered_count=0 root=output skipped=29