MiniMax H3 i2v output is a grid of independently-decoded square tiles

### Custom Node Testing

- [x] I have tried disabling custom nodes and the issue persists (see [how to disable custom nodes](https://docs.comfy.org/troubleshooting/custom-node-issues#step-1%3A-test-with-all-custom-nodes-disabled) if you need help)

### Expected Behavior

Using the official MiniMax H3 Image-to-Video template from workflow_templates. Generation runs fine and produces a video, but every frame is a regular grid of small square tiles, each decoded on its own with its own texture and visible seams between them. It looks like the latent was split into spatial patches and never stitched back together. Not low-res pixelation, not a VAE-file mismatch.

Reproduction

Load the official video_minimax_h3_i2v template.
Download the models the template links (all from Comfy-Org/MiniMax-H3):
diffusion_models: minimax_h3_fl2va_pruned_int8_convrot.safetensors
text_encoders: qwen3vl_32b_minimax_h3_nvfp4_awq.safetensors (loaded as minimax type)
vae: minimax_h3_video_vae_fp16.safetensors
vae (audio): minimax_h3_audio_vae_fp32.safetensors
Set first_frame, resolution 1344x768, duration 5s.
Queue.
Inspect any output frame.
Expected Continuous decoded frames.

Actual Full-frame grid of decoded square patches with seams (see attached frame).

Node chain (as wired in the template, no tiler node present) UNET → CLIP → VAELoader(video_vae_fp16) → MiniMaxH3ImageToVideo → SamplerCustomAdvanced(res_multistep + simple, 20 steps, CFG 1) → VAEDecode(video_vae_fp16) → VAEDecodeAudio(audio_vae_fp32) → CreateVideo → SaveVideo

The video VAE used for decode is minimax_h3_video_vae_fp16 (correct, not the audio VAE). There is no video-tiler node in this template, which suggests H3 latents are spatially patchified and the official VAE wrapper is not stitching the patches.

What I checked

Video VAE is minimax_h3_video_vae_fp16 (official), not the audio VAE.
Not using any third-party H3 TAE (Kijai repo has no weights anyway).
fp16 VAE loads without error.
Resolution is within H3 limits (1344x768, 768px short edge).
Environment

GPU: RTX 4090 (24 GB)
ComfyUI: comfy-core 0.32.0, frontend 1.48.7 (Windows 11)
Extra nodes in chain: SpectrumApplyMiniMaxH3 (comfyui-spectrum-minimax-h3), ModelAttentionBackend set to "comfy kitchen attention"
Questions

Is this a known issue with the official minimax_h3_video_vae_fp16 wrapper when decoding H3's patchified latents?
Does the int8_convrot diffusion model cause it, or does the bf16 model decode cleanly?
Is a tiling/stitching step expected that the template is missing, or should VAEDecode handle it internally?
Frame attached.

### Actual Behavior

https://pastebin.com/irRTPK27

### Steps to Reproduce

### Debug Logs

```powershell
## ComfyUI-Manager: installing dependencies done.
** ComfyUI startup time: 2026-08-12 14:42:55.443
** Platform: Windows
** Python version: 3.12.10 (tags/v3.12.10:0cc8128, Apr 8 2025, 12:21:36) [MSC v.1943 64 bit (AMD64)]
** Python executable: F:\ComfyUI-Easy-Install\python_embeded\python.exe
** ComfyUI Path: F:\ComfyUI-Easy-Install\ComfyUI
** ComfyUI Base Folder Path: F:\ComfyUI-Easy-Install\ComfyUI
** User directory: F:\ComfyUI-Easy-Install\ComfyUI\user
** ComfyUI-Manager config path: F:\ComfyUI-Easy-Install\ComfyUI\user\__manager\config.ini
** Log path: F:\ComfyUI-Easy-Install\ComfyUI\user\comfyui.log
[INFO] CausalLM ModelOutput docstring patch skipped: Qwen VL ModelOutput docstrings resolve loss/logits without docstring errors (upstream fixed — patch not installed)
[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_', 'convrot_w4a4_linear', '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', 'gemv_awq_w4a16', 'int8_linear', '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', 'stochastic_rounding_fp8', 'w4a8_int8_linear']}
[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 hip: {'available': False, 'disabled': False, 'unavailable_reason': 'PyTorch ROCm/HIP runtime not available', 'capabilities': []}
[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', 'gemv_awq_w4a16', 'int8_linear', '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_mxfp8', 'scaled_mm_nvfp4', 'scaled_mm_svdquant_w4a4', 'stochastic_rounding_fp8', 'w4a8_int8_linear']}
[INFO] Checkpoint files will always be loaded safely.
[INFO] Total VRAM 24564 MB, total RAM 65059 MB
[INFO] pytorch version: 2.13.0+cu130
[INFO] Set vram state to: NORMAL_VRAM
[INFO] Device: cuda:0 NVIDIA GeForce RTX 4090 : cudaMallocAsync
[INFO] Using async weight offloading with 2 streams
[INFO] Enabled pinned memory 26023.0
[INFO] Using pytorch attention
[INFO] apply_rotary_emb compat skipped: ComfyUI already exports apply_rotary_emb
[INFO]
Prestartup times for custom nodes:
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-marigold
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\rgthree-comfy
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-Easy-Use
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-grounding
[INFO] 5.6 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-manager
[INFO] 20.6 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-QwenImageLoraLoader
[INFO]
[INFO] Python version: 3.12.10 (tags/v3.12.10:0cc8128, Apr 8 2025, 12:21:36) [MSC v.1943 64 bit (AMD64)]
[INFO] ComfyUI version: 0.32.0
[INFO] comfy-aimdo version: 0.4.13
[INFO] comfy-kitchen version: 0.2.30
[WARNING] Dynamic vram disabled with argument. If you have any issues with dynamic vram enabled please give us a detailed reports as this argument will be removed soon. If you use gguf we recommend keeping dynamic vram enabled and using native ComfyUI model formats instead. ComfyUI native formats like fp8 will be faster even if they are larger than your memory.
[INFO] comfyui-frontend-package version: 1.48.7
[INFO] comfyui-workflow-templates version: 0.11.39
[INFO] comfyui-embedded-docs version: 0.5.9
[INFO] comfy-kitchen version: 0.2.30
[INFO] comfy-aimdo version: 0.4.13
[INFO] [Prompt Server] web root: F:\ComfyUI-Easy-Install\python_embeded\Lib\site-packages\comfyui_frontend_package\static
[INFO] Asset seeder disabled
[INFO] No OpenGL_accelerate module loaded: No module named 'OpenGL_accelerate'
[AymenNode] 📂 Chargement depuis : F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\Aymen_Exif_Compare_V1_X
[AymenNode] ✅ piexif OK
[AymenNode] ✅ aymen_nodes.py chargé avec succès
[AymenNode] ✅ Node enregistré : Aymen_Exif_Compare_V1_X
[F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfy-mtb] | INFO -> loaded 103 nodes successfuly
[F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfy-mtb] | INFO -> Some nodes (5) could not be loaded. This can be ignored, but go to http://127.0.0.1:8188/mtb if you want more information.
Error:
[WinError 1314] Le client ne dispose pas d’un privilège nécessaire: 'F:\\ComfyUI-Easy-Install\\ComfyUI\\custom_nodes\\ComfyLiterals\\js' -> 'F:\\ComfyUI-Easy-Install\\ComfyUI\\web\\extensions\\ComfyLiterals'
Failed to create symlink to F:\ComfyUI-Easy-Install\ComfyUI\web\extensions\ComfyLiterals. Please copy the folder manually.
Source: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyLiterals\js
Target: F:\ComfyUI-Easy-Install\ComfyUI\web\extensions\ComfyLiterals
[INFO] Imported AddPaddingAdvanced successfully
[INFO] Imported AddPaddingBase successfully
[INFO] Imported AD_ImageResize successfully
[INFO] Imported AD_MockupMaker successfully
[INFO] Imported AD_PosterMaker successfully
[INFO] Imported AD_PromptSaver successfully
[INFO] Imported ComfyUI-FofrToolkit successfully
[INFO] Imported ComfyUI-ImageCaptioner successfully
[INFO] Imported ComfyUI-imageResize successfully
[INFO] Imported ComfyUI-textAppend successfully
[INFO] Imported imagecreatemask successfully
[INFO] Imported multiline_string successfully
[INFO] Imported AddPaddingAdvanced successfully
[INFO] Imported AddPaddingBase successfully
[INFO] Imported AD_ImageResize successfully
[INFO] Imported AD_MockupMaker successfully
[INFO] Imported AD_PosterMaker successfully
[INFO] Imported AD_PromptSaver successfully
[INFO] Imported ComfyUI-FofrToolkit successfully
[INFO] Imported ComfyUI-ImageCaptioner successfully
[INFO] Imported ComfyUI-imageResize successfully
[INFO] Imported ComfyUI-textAppend successfully
[INFO] Imported imagecreatemask successfully
[INFO] Imported multiline_string successfully

=================================================================
SID Photography Toolkit for ComfyUI
Version: 4.2.0
Author: Siddhartha Lahiri
=================================================================

Dependencies:
[OK] Anthropic Claude API
[OK] OpenAI/Grok/LM Studio API
[OK] YAML config parser
[OK] HTTP library (Ollama)
[OK] Type hints

-----------------------------------------------------------------
Available Nodes:

LLM Providers:
- SID_LLM_API Cloud APIs (Anthropic, OpenAI, Gemini, Grok, Ollama, LM Studio)
- SID_LLM_Local Local models (Florence-2, Moondream, SmolVLM, Phi-3.5, QwenVL)

Prompt Generator:
- SID_ZImagePromptGenerator Z-Image prompts (auto Single-Shot/Agentic)

=================================================================

[custom_nodes.comfyui_controlnet_aux] | INFO -> Using ckpts path: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui_controlnet_aux\ckpts
[custom_nodes.comfyui_controlnet_aux] | INFO -> Using symlinks: False
[custom_nodes.comfyui_controlnet_aux] | INFO -> Using ort providers: ['CUDAExecutionProvider', 'DirectMLExecutionProvider', 'OpenVINOExecutionProvider', 'ROCMExecutionProvider', 'CPUExecutionProvider', 'CoreMLExecutionProvider']
F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui_controlnet_aux\node_wrappers\dwpose.py:26: UserWarning: DWPose: Onnxruntime not found or doesn't come with acceleration providers, switch to OpenCV with CPU device. DWPose might run very slowly
warnings.warn("DWPose: Onnxruntime not found or doesn't come with acceleration providers, switch to OpenCV with CPU device. DWPose might run very slowly")
Adding F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes to sys.path
Could not find efficiency nodes
[comfyui_controlnet_aux] | INFO -> Using ckpts path: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui_controlnet_aux\ckpts
[comfyui_controlnet_aux] | INFO -> Using symlinks: False
[comfyui_controlnet_aux] | INFO -> Using ort providers: ['CUDAExecutionProvider', 'DirectMLExecutionProvider', 'OpenVINOExecutionProvider', 'ROCMExecutionProvider', 'CPUExecutionProvider', 'CoreMLExecutionProvider']
Loaded comfyui_controlnet_aux nodes from F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui_controlnet_aux
Could not find AdvancedControlNet nodes
Could not find AnimateDiff nodes
Could not find IPAdapter nodes
Could not find VideoHelperSuite nodes
[INFO] ### Loading: ComfyUI-Impact-Pack (V8.28.3)
[INFO] ### Loading: ComfyUI-Impact-Pack (V8.28.3)
Loaded ImpactPack nodes from F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-impact-pack
[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] [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] [ClipProj] graph watch installed
[INFO] [ClipProj] unload hook installed
[Crystools INFO] Crystools version: 1.27.4
[Crystools INFO] Platform release: 11
[Crystools INFO] JETSON: Not detected.
[Crystools INFO] CPU: AMD Ryzen 9 7950X3D 16-Core Processor - Arch: AMD64 - OS: Windows 11
[Crystools INFO] pynvml (NVIDIA) initialized.
[Crystools INFO] GPU/s:
[Crystools INFO] 0) NVIDIA GeForce RTX 4090
[Crystools INFO] NVIDIA Driver: 595.97
[Qwen3VL] ℹ️ 未找到 custom_models.json，跳过自定义模型
[Qwen3VL] ℹ️ 未找到 custom_models.json，跳过自定义模型
✅ 配置文件加载成功！
📊 共找到 20 个模型

✅ 新模型 'Huihui-Qwen3-VL-4B-Instruct-Abliterated' 配置成功！
📦 Repo ID: fireicewolf/Huihui-Qwen3-VL-4B-Instruct-abliterated
🌐 来源: modelscope
💾 显存需求: {'full': 6.0, '8bit': 3.5, '4bit': 2.0}
⚠️ 警告: 此模型已移除安全过滤，可能生成敏感内容。仅用于研究和测试环境。

📋 所有可用模型:
1. Qwen3-VL-2B-Instruct (来源: huggingface)
2. Qwen3-VL-2B-Thinking (来源: huggingface)
3. Qwen3-VL-2B-Instruct-FP8 (来源: huggingface)
4. Qwen3-VL-2B-Thinking-FP8 (来源: huggingface)
5. Qwen3-VL-4B-Instruct (来源: huggingface)
6. Qwen3-VL-4B-Thinking (来源: huggingface)
7. Qwen3-VL-4B-Instruct-FP8 (来源: huggingface)
8. Qwen3-VL-4B-Thinking-FP8 (来源: huggingface)
9. Qwen3-VL-8B-Instruct (来源: huggingface)
10. Qwen3-VL-8B-Thinking (来源: huggingface)
11. Qwen3-VL-8B-Instruct-FP8 (来源: huggingface)
12. Qwen3-VL-8B-Thinking-FP8 (来源: huggingface)
13. Qwen3-VL-32B-Instruct (来源: huggingface)
14. Qwen3-VL-32B-Thinking (来源: huggingface)
15. Qwen3-VL-32B-Instruct-FP8 (来源: huggingface)
16. Qwen3-VL-32B-Thinking-FP8 (来源: huggingface)
17. Qwen2.5-VL-3B-Instruct (来源: huggingface)
18. Qwen2.5-VL-7B-Instruct (来源: huggingface)
19. Huihui-Qwen3-VL-4B-Instruct-Abliterated (来源: modelscope)
20. Huihui-Qwen3-VL-8B-Instruct-Abliterated (来源: huggingface)
[ComfyUI-Darkroom] registered HTTP routes: /darkroom/list_dir, /darkroom/mkdir, /darkroom/camera_looks, /darkroom/pins (GET/POST/DELETE)
[ComfyUI-Darkroom] Loaded — 59 nodes (Film Emulation + Camera Raw + Color Grading + Lens & Optics + LUT Pipeline + RAW Pipeline w/ DCP + Spectral Film Stock + Scopes + Color Match + CMYK Print)
Multiple distributions found for package optimum. Picked distribution: optimum
[WARNING] Skipping import of cpp extensions due to incompatible torch version 2.13.0+cu130 for torchao version 0.14.1 Please see https://github.com/pytorch/ao/issues/2919 for more info
[ComfyUI-Easy-Use] server: v1.3.7 Loaded
[ComfyUI-Easy-Use] web root: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-Easy-Use\web_version/v2 Loaded
[FishAudioS2] Models folder registered: F:\ComfyUI-Easy-Install\ComfyUI\models\fishaudioS2
[FishAudioS2] Registered 4 nodes (v0.5.3): Fish S2 TTS, Fish S2 Voice Clone TTS, Fish S2 Multi-Speaker TTS, Fish S2 Multi-Speaker Split TTS
[INFO] ComfyUI-GGUF: Allowing full torch compile
[ComfyUI-Grounding] Initializing custom node...
[ComfyUI-Grounding] Initializing...
[ComfyUI-Grounding] Modern ComfyUI detected - using automatic web extension serving
[ComfyUI-Grounding] [OK] Initialization script completed
[ComfyUI-Grounding] Loading nodes module...
[ComfyUI-Grounding] [OK] Imported basic dependencies (torch, numpy, etc.)
[ComfyUI-Grounding] [OK] Imported config (MODEL_REGISTRY, MASK_MODEL_REGISTRY)
[ComfyUI-Grounding] [OK] Imported utils.cache
[ComfyUI-Grounding] [OK] Imported utils (draw_boxes, boxes_to_masks)
[ComfyUI-Grounding] [OK] Imported grounding_dino
[ComfyUI-Grounding] [OK] Imported owlv2
[ComfyUI-Grounding] [OK] Imported florence2
[ComfyUI-Grounding] [OK] Imported yolo_world
[ComfyUI-Grounding] [OK] Imported florence2_seg
[ComfyUI-Grounding] [OK] Imported sa2va
[ComfyUI-Grounding] [OK] Imported sam2
[ComfyUI-Grounding] [OK] Imported visualizer
[ComfyUI-Grounding] [OK] All modules imported successfully
[ComfyUI-Grounding] [OK] Node classes imported successfully
[ComfyUI-Grounding] [OK] Loaded successfully!
[INFO] ### Loading: ComfyUI-Impact-Pack (V8.28.3)
[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] ### Loading: ComfyUI-Inspire-Pack (V1.23)
[InstaMockup] loaded - 1 node(s)
[krea2edit] nodes v1.2.5 loaded
[llama-cpp-python] loaded bundled OpenMP runtime: F:\ComfyUI-Easy-Install\python_embeded\Lib\site-packages\llama_cpp\lib\libomp140.x86_64.dll
[llama-cpp-python].find_library: loaded library from F:\ComfyUI-Easy-Install\python_embeded\Lib\site-packages\llama_cpp\lib\ggml-base.dll
[llama-cpp-python].find_library: loaded library from F:\ComfyUI-Easy-Install\python_embeded\Lib\site-packages\llama_cpp\lib\ggml.dll
[llama-cpp-python].find_library: loaded library from F:\ComfyUI-Easy-Install\python_embeded\Lib\site-packages\llama_cpp\lib\llama.dll
[llama-cpp_vlm] Model cleanup hook applied!
[INFO] [LoRA-Manager] Found LoRA roots:
- F:/ComfyUI-Easy-Install/ComfyUI/models/loras
- F:/Soft and Tools/Soft and Tools/Ai/Comfyui3/models/loras
[INFO] [LoRA-Manager] Found checkpoint roots:
- F:/ComfyUI-Easy-Install/ComfyUI/models/checkpoints
- F:/ComfyUI-Easy-Install/ComfyUI/models/diffusion_models
- F:/ComfyUI-Easy-Install/ComfyUI/models/unet
- F:/Soft and Tools/Soft and Tools/Ai/Comfyui3/models/checkpoints
- F:/Soft and Tools/Soft and Tools/Ai/Comfyui3/models/diffusion_models
- F:/Soft and Tools/Soft and Tools/Ai/Comfyui3/models/unet
[INFO] [LoRA-Manager] Found embedding roots:
- F:/ComfyUI-Easy-Install/ComfyUI/models/embeddings
- F:/Soft and Tools/Soft and Tools/Ai/Comfyui3/models/embeddings
[INFO] [LoRA-Manager] Symlink cache loaded with 0 mappings
[INFO] [LoRA-Manager] Symlink mappings restored from cache in 1.63 ms
[INFO] [LoRA-Manager] Updated 'comfyui' library with current folder paths
[INFO] [LoRA-Manager] Detected async ComfyUI execution, installing async metadata hooks
[INFO] [LoRA-Manager] Metadata collection hooks installed for runtime values
[INFO] [LoRA-Manager] ComfyUI Metadata Collector initialized
[INFO] [LoRA-Manager] Example images path:
[INFO] [LoRA-Manager] Added static route for locales: /locales -> F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-lora-manager\locales
[INFO] ### Loading: ComfyUI-Manager (V3.41)
[INFO] [ComfyUI-Manager] network_mode: public
[INFO] [ComfyUI-Manager] ComfyUI per-queue preview override detected (PR #11261). Manager's preview method feature is disabled. Use ComfyUI's --preview-method CLI option or 'Settings > Execution > Live preview method'.
[INFO] ### ComfyUI Version: v0.32.0-1-g27bca654 | Released on '2026-08-11'
[INFO] Total VRAM 24564 MB, total RAM 65059 MB
[INFO] pytorch version: 2.13.0+cu130
[INFO] Set vram state to: NORMAL_VRAM
[INFO] Device: cuda:0 NVIDIA GeForce RTX 4090 : cudaMallocAsync
[INFO] Using async weight offloading with 2 streams
[INFO] Enabled pinned memory 26023.0
[INFO] [ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/alter-list.json
[INFO] [ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/github-stats.json
[INFO] [ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/model-list.json
[INFO] [ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/extension-node-map.json
[INFO] [ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/custom-node-list.json
vtracer module found. SVG conversion will use vtracer.
============================================================
ComfyUI-OminiKontext loaded successfully!
Version: 1.0.0
Available nodes: 13
- Omini Kontext Pipeline Loader
- Omini Kontext Pipeline
- Omini Kontext Image Scale
- Omini Kontext Split Pipeline Loader
- Omini Kontext LoRA Loader
- Omini Kontext LoRA Unload
- Omini Kontext LoRA Merge
- Omini Kontext Image Encoder
- Omini Kontext Text Encoder
- Omini Kontext Reference Encoder
- Omini Kontext Latent Combiner
- Omini Kontext Latent Decoder
- Omini Kontext Latent Visualizer
============================================================
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Pixaroma v1.4.103 | 72 nodes Loaded
ComfyUI Tutorials: https://www.youtube.com/@pixaroma
This is a notice, not an error. All Pixaroma nodes work in both Classic and Nodes 2.0 mode.
If something looks off right after switching the Node UI mode, hard-refresh the page (Ctrl+Shift+R).
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
[Prompt Rewriter] Job Object created - llama-server will be killed on console close
[Prompt Rewriter] Job Object created - llama-server will be killed on console close
[Prompt Rewriter] Console close handler registered
[INFO] ======================================== ComfyUI-QwenImageLoraLoader Initialization ========================================
[INFO] apply_rotary_emb compat skipped: ComfyUI already exports apply_rotary_emb
[WARNING] nunchaku.models.linear.SVDQW4A4Linear not importable; lazy-Linear patch skipped.
[ERROR] Error importing/applying Nunchaku monkey patch: No module named 'nunchaku'
[INFO] ControlNet node loaded successfully
[INFO] ============================================================================================================================
[QwenImageWanBridge] Loaded Qwen2.5-VL nodes using ComfyUI's CLIP
[QwenImageWanBridge] Loaded Advanced Encoder with weighted resolution control
[QwenImageWanBridge] Loaded Qwen VL helper nodes (3 nodes)
[QwenImageWanBridge] Loaded Qwen Image Batch node
[QwenImageWanBridge] FEATURE: Aspect-ratio preserving batching with v2.6.1 scaling modes
[QwenImageWanBridge] Loaded Qwen-to-Wan Video Bridge nodes (3 nodes)
[QwenImageWanBridge] FEATURE: Image-to-video bridge with first frame conditioning
[QwenImageWanBridge] Loaded Qwen-to-ChronoEdit Bridge node
[QwenImageWanBridge] FEATURE: Qwen image editing → ChronoEdit video animation
[QwenImageWanBridge] Loaded mask-based inpainting nodes (2 nodes)
[QwenImageWanBridge] FEATURE: Mask-based spatial editing with diffusers patterns
[QwenImageWanBridge] Loaded template builder nodes
[QwenImageWanBridge] Loaded EliGen Entity Control nodes (2 nodes)
[QwenImageWanBridge] FEATURE: Precise spatial control with masks and per-region prompts
[QwenImageWanBridge] Loaded Token Analysis nodes (3 nodes)
[QwenImageWanBridge] Failed to load Debug Controller: No module named 'debug_patch'
[QwenImageWanBridge] Loaded Experimental Smart Crop node
[QwenImageWanBridge] EXPERIMENTAL: Intelligent face cropping with VLM detection support
[QwenImageWanBridge] Total nodes loaded: 22
[QwenImageWanBridge] Loaded HunyuanVideo 1.5 loader nodes (3 nodes)
[QwenImageWanBridge] FEATURE: Dual CLIP loader (Qwen2.5-VL + byT5)
[QwenImageWanBridge] Loaded HunyuanVideo text encoder
[QwenImageWanBridge] API: /api/hunyuan_video_templates endpoint registered
[QwenImageWanBridge] Loaded HunyuanVideo prompt expander
[QwenImageWanBridge] FEATURE: Lightweight prompt expansion using loaded Qwen2.5-VL
[QwenImageWanBridge] Loaded Z-Image encoder nodes (4 nodes)
[QwenImageWanBridge] API: /api/z_image_templates endpoint registered
[QwenImageWanBridge] Loaded LLM Output Parser node
[QwenImageWanBridge] FEATURE: Parse JSON/YAML/text LLM output to Z-Image encoder inputs
[QwenImageWanBridge] Loaded Z-Image Wan VAE Decode (EXPERIMENTAL - architecture mismatch)
[QwenImageWanBridge] Loaded Template Influence Analyzer (experimental)
[ComfyUI-RMBG] v3.1.0 | 43 nodes Loaded
Workspace manager - Openning file hash dict
### Loading: Workspace Manager (V1.0.0)
[Z-Image-Turbo] Registered samplers: ['euler_flow', 'euler_flow_alt', 'euler_flow_verbose', 'midpoint_flow', 'heun_flow']
Python version is above 3.10, patching the collections module.
------------------------------------------
Comfyroll Studio v1.76 : 175 Nodes Loaded
------------------------------------------
** For changes, please see patch notes at https://github.com/Suzie1/ComfyUI_Comfyroll_CustomNodes/blob/main/Patch_Notes.md
** For help, please see the wiki at https://github.com/Suzie1/ComfyUI_Comfyroll_CustomNodes/wiki
------------------------------------------
[F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui_controlnet_aux] | INFO -> Using ckpts path: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui_controlnet_aux\ckpts
[F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui_controlnet_aux] | INFO -> Using symlinks: False
[F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui_controlnet_aux] | INFO -> Using ort providers: ['CUDAExecutionProvider', 'DirectMLExecutionProvider', 'OpenVINOExecutionProvider', 'ROCMExecutionProvider', 'CPUExecutionProvider', 'CoreMLExecutionProvider']
[INFO] MiniMax H3 Director HTTP routes registered
✅ ComfyUI Reve API Integration loaded successfully
[tinyterraNodes] Loaded

Initializing ControlAltAI Nodes
[RemoveBackground_SET] Registering 15 node(s) for version 1.0.0.
[INFO] Using pytorch attention
(RES4LYF) Init
(RES4LYF) Importing beta samplers.
(RES4LYF) Importing legacy samplers.

[rgthree-comfy] Loaded 48 magnificent nodes. 🎉

[rgthree-comfy] ComfyUI's new Node 2.0 rendering may be incompatible with some rgthree-comfy nodes and features, breaking some rendering as well as losing the ability to access a node's properties (a vital part of many nodes). It also appears to run MUCH more slowly spiking CPU usage and causing jankiness and unresponsiveness, especially with large workflows. Personally I am not planning to use the new Nodes 2.0 and, unfortunately, am not able to invest the time to investigate and overhaul rgthree-comfy where needed. If you have issues when Nodes 2.0 is enabled, I'd urge you to switch it off as well and join me in hoping ComfyUI is not planning to deprecate the existing, stable canvas rendering all together.

✅ Đã cài aria2c.
Đã cập nhật F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\sdvn_comfy_node\config.json
⚡ SeedVR2 optimizations check: SageAttention ✅ | Flash Attention ✅ | Triton ✅
🔧 Conv3d workaround active: PyTorch 2.13.0, cuDNN 92000 (fixing VAE 3x memory bug)
📊 Initial CUDA memory: 22.38GB free / 23.99GB total
[StyleRef] v1.0.6 — 5 nodes registered
[VibeVoice] SageAttention available for acceleration
[VibeVoice] Using embedded VibeVoice (MIT licensed)
[VibeVoice] VibeVoice nodes registered successfully
WAS Node Suite: OpenCV Python FFMPEG support is enabled
WAS Node Suite Warning: `ffmpeg_bin_path` is not set in `F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\was-node-suite-comfyui\was_suite_config.json` config file. Will attempt to use system ffmpeg binaries if available.
WAS Node Suite: Finished. Loaded 220 nodes successfully.

"The harder I work, the luckier I get." - Samuel Goldwyn

[WARNING] Traceback (most recent call last):
File "F:\ComfyUI-Easy-Install\ComfyUI\nodes.py", line 2260, in load_custom_node
module_spec.loader.exec_module(module)
File "", line 995, in exec_module
File "", line 1132, in get_code
File "", line 1190, in get_data
FileNotFoundError: [Errno 2] No such file or directory: 'F:\\ComfyUI-Easy-Install\\ComfyUI\\custom_nodes\\_disabled\\__init__.py'

[WARNING] Cannot import F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\_disabled module for custom nodes: [Errno 2] No such file or directory: 'F:\\ComfyUI-Easy-Install\\ComfyUI\\custom_nodes\\_disabled\\__init__.py'
[INFO]
Import times for custom nodes:
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\websocket_image_save.py
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\canvas_tab
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-clark-swin2sr
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui_caption_cleaner
[INFO] 0.0 seconds (IMPORT FAILED): F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\_disabled
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui_caption_master
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\GemstoneRetouchNode
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui_batch_caption_cleaner
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui_project_dataset_manager
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI_AdvancedRefluxControl
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui_dataset_generation_flux2
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI_Noise
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\scale-image-to-total-pixels-advanced
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI_LinkFX
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui_nvidia_rtx_nodes
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\Sphere-Light-Render-ComfyUI
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-EsesImageCompare
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\gemini-api
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui_ttp_toolset
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-comfysketch
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-AymenBadr-Ultimate
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-mxtoolkit
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-seamless-tiling
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUi-TextEncodeEditAdvanced
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\referencechain
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui_gemstone_tools
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI_Text_Translation
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\wywywywy-pause
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-Krea2-Ostris-Edit
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\openrouter_node
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-ZImageTurbo-FlowSampler
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\GF_nodes
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-krea2edit
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-detail-daemon
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ai_toolkit_trainer
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\seedvarianceenhancer
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-inpaint-cropandstitch
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-Prompt_Helper
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyUI_FrequencySeparation_RGB-HSV
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-BRIA_AI-RMBG
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\janus-pro
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI_KleinTiledUpscaler
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-CapitanFlowMatch
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\cg-use-everywhere
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui_steudio
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-SCAIL-Pose
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyLiterals
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI_Searge_LLM
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui_style_composer
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-Krea2T-Enhancer
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\qwen-image-integrated-ksampler
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\Comfyui-flux2klein-Lora-loader
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI_Qwen3-VL-Instruct
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\omini-kontext
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-sg-llama-cpp
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\eacloudnodes
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-angelo
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-InstaMockup
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-Florence2
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-External-Lora-Loader
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\qwen3-vl-nsfw
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\z-image-turbo
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-ClipProj
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-RuiquNodes
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-LG_SamplingUtils
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\erosdiffusion-eulerflowmatchingdiscretescheduler
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ControlFlowUtils
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI_Reve-API
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-PixelDriftFix
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-lama-remover
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-omnigen
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\Aymen_Exif_Compare_V2Arena
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-GGUF
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-MelBandRoFormer
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-Prompt-Manager
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-WanAnimatePreprocess
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-inpaint-nodes
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\Comfyui-Resolution-Master
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\Aymen_Exif_Compare_V1_X
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\styleref-comfyui
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-Flux2Klein-Enhancer
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui_essentials
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-ak-pack
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\VibeVoice-ComfyUI
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\Comfy_ForEach
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-enricos-nodes
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-video-tiler
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\qwen2_5-vl
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-qwen35
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-DyPE
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-impact-pack
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-QwenVL
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-custom-scripts
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\controlaltai-nodes
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-QwenImageLoraLoader
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-AI-Photography-Toolkit
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI_Sonic
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-sam2
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-Qwen3-VL
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI_Simple_Qwen3-VL-gguf
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-grounding
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-Spectrum-MiniMax-H3
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\rgthree-comfy
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui_lg_tools
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-QwenImageWanBridge
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI_MiniMaxH3_Director
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-mickmumpitz-nodes
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui_caption_this
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\no8d-controls
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui_birefnet_ll
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui_controlnet_aux
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\basic_data_handling
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI_Comfyroll_CustomNodes
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui_mienodes
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-kjnodes
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-inspire-pack
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-videohelpersuite
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\kaytool
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-llama-cpp
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui_tinyterranodes
[INFO] 0.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-Omini-Kontext
[INFO] 0.1 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-llm-session
[INFO] 0.1 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-ToSVG
[INFO] 0.1 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-LTXVideo
[INFO] 0.1 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\remove-background
[INFO] 0.1 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-PIX_Nvidia_RTX_Deblur_Denoise
[INFO] 0.1 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-dapao-tool
[INFO] 0.1 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-JewelryTools
[INFO] 0.1 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-workspace-manager
[INFO] 0.1 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-anyline
[INFO] 0.1 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-Pixaroma
[INFO] 0.1 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui_layerstyle
[INFO] 0.1 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-Qwen-VL-API
[INFO] 0.1 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-logicutils
[INFO] 0.1 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-marigold
[INFO] 0.2 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\seedvr2_videoupscaler
[INFO] 0.2 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\RES4LYF
[INFO] 0.2 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-lora-manager
[INFO] 0.2 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\sdvn_comfy_node
[INFO] 0.2 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-ollama
[INFO] 0.3 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-fish-audio-s2
[INFO] 0.3 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-Crystools
[INFO] 0.3 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-rmbg
[INFO] 0.3 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-art-venture
[INFO] 0.4 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-propost
[INFO] 0.4 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\Artha-Gemini
[INFO] 0.4 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-darkroom
[INFO] 0.5 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-WanVideoWrapper
[INFO] 0.5 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-advancedliveportrait
[INFO] 0.5 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-kokoro
[INFO] 0.6 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-itools
[INFO] 1.0 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-Addoor
[INFO] 1.2 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfy-mtb
[INFO] 1.2 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\was-node-suite-comfyui
[INFO] 1.4 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-manager
[INFO] 1.4 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-ollamagemini
[INFO] 2.4 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-GeekyRemB
[INFO] 2.6 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-gemini
[INFO] 2.9 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-easy-sam3
[INFO] 4.7 seconds: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-Easy-Use
[INFO]
[INFO] Context impl SQLiteImpl.
[INFO] Will assume non-transactional DDL.
[INFO] Using RAM pressure cache.
[INFO] [LoRA-Manager] Deduplicate: removed 0 history rows, 0 queue rows, 0 orphaned queue rows
[INFO] Starting server

[INFO] To see the GUI go to: http://127.0.0.1:8188
[INFO] [LoRA-Manager] Recipe cache miss: performing full directory scan
[WARNING] [LoRA-Manager] Duplicate filename conflict detected: 2 lora filename(s) are shared by 6 files total, causing ambiguity in Lora resolution. Open the Doctor panel to resolve one-click.
[INFO] [LoRA-Manager] Recipe cache initialized in 0.08 seconds. Found 0 recipes
[INFO] [LoRA-Manager] Lora cache hydrated from persisted snapshot with 181 models
[INFO] [LoRA-Manager] Checkpoint cache hydrated from persisted snapshot with 25 models
[INFO] [LoRA-Manager] Embedding cache hydrated from persisted snapshot with 1 models
[INFO] [LoRA-Manager] FTS index validated, reusing existing index with 0 recipes
[WARNING] [DEPRECATION WARNING] Detected import of deprecated legacy API: /scripts/ui.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.
[WARNING] [DEPRECATION WARNING] Detected import of deprecated legacy API: /extensions/core/clipspace.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.
[WARNING] [DEPRECATION WARNING] Detected import of deprecated legacy API: /scripts/ui/components/buttonGroup.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.
[WARNING] [DEPRECATION WARNING] Detected import of deprecated legacy API: /extensions/core/groupNode.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.
[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.
[F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfy-mtb] | INFO -> Found multiple match, we will pick the last F:\ComfyUI-Easy-Install\ComfyUI\models\upscale_models
['F:\\Soft and Tools\\Soft and Tools\\Ai\\Comfyui3\\models\\upscale_models', 'F:\\ComfyUI-Easy-Install\\ComfyUI\\models\\upscale_models']
[ERROR] [Inspire Pack] IPAdapterPlus is not installed.
Gemini models fetched: 52 models available
Gemini models fetched: 52 models available
Error fetching Ollama models: HTTPConnectionPool(host='localhost', port=11434): Max retries exceeded with url: /api/tags (Caused by NewConnectionError("HTTPConnection(host='localhost', port=11434): Failed to establish a new connection: [WinError 10061] Aucune connexion n’a pu être établie car l’ordinateur cible l’a expressément refusée"))
Error fetching Ollama models: HTTPConnectionPool(host='localhost', port=11434): Max retries exceeded with url: /api/tags (Caused by NewConnectionError("HTTPConnection(host='localhost', port=11434): Failed to establish a new connection: [WinError 10061] Aucune connexion n’a pu être établie car l’ordinateur cible l’a expressément refusée"))
Final Gemini image models list: ['gemini-2.5-flash-image', 'gemini-3-pro-image-preview', 'gemini-3-pro-image', 'gemini-3.1-flash-image-preview', 'gemini-3.1-flash-image', 'gemini-3.1-flash-lite-image', 'imagen-4.0-generate-001', 'imagen-4.0-ultra-generate-001', 'imagen-4.0-fast-generate-001', 'gemini-2.5-flash-image-preview']
Final Gemini image models list: ['gemini-2.5-flash-image', 'gemini-3-pro-image-preview', 'gemini-3-pro-image', 'gemini-3.1-flash-image-preview', 'gemini-3.1-flash-image', 'gemini-3.1-flash-lite-image', 'imagen-4.0-generate-001', 'imagen-4.0-ultra-generate-001', 'imagen-4.0-fast-generate-001', 'gemini-2.5-flash-image-preview']
[Qwen3VL Config] ✓ Config loaded from: F:\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-Qwen3-VL\Qwen3-VL-config.json
[INFO] Loaded 9 templates
[VibeVoice] No valid models found in vibevoice directory
[VibeVoice] Please download models to: F:\Soft and Tools\Soft and Tools\Ai\Comfyui3\models\vibevoice
[VibeVoice] No LoRA adapters found in ComfyUI/models/vibevoice/loras
[WARNING] [DEPRECATION WARNING] Detected import of deprecated legacy API: /scripts/ui/components/button.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] got prompt
[INFO] VAE load device: cuda:0, offload device: cpu, dtype: torch.float32
[INFO] VAE load device: cuda:0, offload device: cpu, dtype: torch.float16
[INFO] Found quantization metadata version 1
[INFO] Using MixedPrecisionOps for text encoder
[INFO] CLIP/text encoder model load device: cuda:0, offload device: cpu, current: cpu, dtype: torch.float16
[INFO] Requested to load MiniMaxH3TEModel_
[INFO] loaded completely; 21418.67 MB usable, 14960.20 MB loaded, full load: True
[INFO] Found quantization metadata version 1
[INFO] Detected mixed precision quantization
[INFO] Using mixed precision operations
[INFO] Native ops: asym_w4a8_int8, convrot_w4a4, float8_e4m3fn, int8_tensorwise, float8_e5m2 , emulated ops: nvfp4, mxfp8
[INFO] model weight dtype torch.bfloat16, manual cast: torch.bfloat16
[INFO] model_type FLOW_AV
[WARNING] Warning: TAESD previews enabled, but could not find models/vae_approx/None
[INFO] Requested to load MiniMaxH3
[INFO] loaded completely; 20253.73 MB usable, 19996.14 MB loaded, full load: True
100%|██████████████████████████████████████████████████████████████████████████████████| 20/20 [00:38<00:00, 1.94s/it]
[INFO] Requested to load MiniMaxH3AudioVAE
[INFO] Unloaded partially: 360.05 MB freed, 19636.13 MB remains loaded, 257.27 MB buffer reserved, lowvram patches: 0
[INFO] loaded completely; 855.50 MB usable, 577.08 MB loaded, full load: True
[INFO] Requested to load MiniMaxH3VideoVAE
[INFO] Unloaded partially: 5587.85 MB freed, 14048.28 MB remains loaded, 661.52 MB buffer reserved, lowvram patches: 0
[INFO] loaded completely; 6741.17 MB usable, 4966.19 MB loaded, full load: True
[Pixaroma] Save Mp4 [save] — writing 73 frames @ 24fps (480x864, crf=19, yuv420p, +audio) -> Video_00007.mp4
[Pixaroma] Save Mp4 — saved F:\ComfyUI-Easy-Install\ComfyUI\output\Video_00007.mp4
[INFO] Prompt executed in 89.27 seconds
[INFO] [LoRA-Manager] Calculating hash for checkpoint 'minimax_h3_fl2va_pruned_int8_convrot' from F:/Soft and Tools/Soft and Tools/Ai/Comfyui3/models/unet/H3/minimax_h3_fl2va_pruned_int8_convrot.safetensors
FETCH ComfyRegistry Data [DONE]
[INFO] [ComfyUI-Manager] default cache updated: https://api.comfy.org/nodes
FETCH DATA from: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/custom-node-list.json [DONE]
[INFO] [ComfyUI-Manager] All startup tasks have been completed.
[INFO] got prompt
[WARNING] Warning: TAESD previews enabled, but could not find models/vae_approx/None
[WARNING] Warning: TAESD previews enabled, but could not find models/vae_approx/None
[INFO] Requested to load MiniMaxH3
[INFO] loaded completely; 20211.73 MB usable, 19996.14 MB loaded, full load: True
100%|██████████████████████████████████████████████████████████████████████████████████| 20/20 [00:38<00:00, 1.93s/it]
[INFO] Requested to load MiniMaxH3AudioVAE
[INFO] loaded partially; 482.87 MB usable, 461.86 MB loaded, 115.02 MB offloaded, 21.01 MB buffer reserved, lowvram patches: 0
[INFO] Requested to load MiniMaxH3VideoVAE
[INFO] Unloaded partially: 5506.78 MB freed, 14489.40 MB remains loaded, 661.52 MB buffer reserved, lowvram patches: 0
[INFO] loaded completely; 6869.16 MB usable, 4966.19 MB loaded, full load: True
[Pixaroma] Save Mp4 [save] — writing 73 frames @ 24fps (480x864, crf=19, yuv420p, +audio) -> Video_00008.mp4
[Pixaroma] Save Mp4 — saved F:\ComfyUI-Easy-Install\ComfyUI\output\Video_00008.mp4
[INFO] Prompt executed in 50.87 seconds
[INFO] got prompt
[WARNING] Warning: TAESD previews enabled, but could not find models/vae_approx/None
[WARNING] Warning: TAESD previews enabled, but could not find models/vae_approx/None
[INFO] Requested to load MiniMaxH3
[INFO] loaded completely; 20209.73 MB usable, 19996.14 MB loaded, full load: True
100%|██████████████████████████████████████████████████████████████████████████████████| 20/20 [00:28<00:00, 1.44s/it]
[INFO] Requested to load MiniMaxH3AudioVAE
[INFO] Unloaded partially: 396.82 MB freed, 19599.36 MB remains loaded, 257.27 MB buffer reserved, lowvram patches: 0
[INFO] loaded completely; 1054.99 MB usable, 577.08 MB loaded, full load: True
[INFO] Requested to load MiniMaxH3VideoVAE
[INFO] Unloaded partially: 5551.08 MB freed, 14048.28 MB remains loaded, 661.52 MB buffer reserved, lowvram patches: 0
[INFO] loaded completely; 6733.17 MB usable, 4966.19 MB loaded, full load: True
[Pixaroma] Save Mp4 [save] — writing 73 frames @ 24fps (480x864, crf=19, yuv420p, +audio) -> Video_00009.mp4
[Pixaroma] Save Mp4 — saved F:\ComfyUI-Easy-Install\ComfyUI\output\Video_00009.mp4
[INFO] Prompt executed in 40.02 seconds
[INFO] got prompt
[WARNING] Warning: TAESD previews enabled, but could not find models/vae_approx/None
[WARNING] Warning: TAESD previews enabled, but could not find models/vae_approx/None
[INFO] Requested to load MiniMaxH3
[INFO] loaded completely; 20209.73 MB usable, 19996.14 MB loaded, full load: True
100%|██████████████████████████████████████████████████████████████████████████████████| 20/20 [00:18<00:00, 1.11it/s]
[WARNING] Warning: TAESD previews enabled, but could not find models/vae_approx/None
[INFO] 0 models unloaded.
[INFO] Unloaded partially: 911.61 MB freed, 19084.57 MB remains loaded, 257.27 MB buffer reserved, lowvram patches: 0
[INFO] Requested to load MiniMaxH3AudioVAE
[INFO] loaded completely; 1506.56 MB usable, 577.08 MB loaded, full load: True
[INFO] Requested to load MiniMaxH3VideoVAE
[INFO] Unloaded partially: 5109.81 MB freed, 13974.76 MB remains loaded, 661.52 MB buffer reserved, lowvram patches: 0
[INFO] loaded completely; 6806.69 MB usable, 4966.19 MB loaded, full load: True
[Pixaroma] Save Mp4 [save] — writing 73 frames @ 24fps (480x864, crf=19, yuv420p, +audio) -> Video_00010.mp4
[Pixaroma] Save Mp4 — saved F:\ComfyUI-Easy-Install\ComfyUI\output\Video_00010.mp4
[INFO] Prompt executed in 33.52 seconds
[INFO] got prompt
[INFO] Requested to load MiniMaxH3TEModel_
[INFO] Unloaded partially: 983.75 MB freed, 3982.74 MB remains loaded, 72.04 MB buffer reserved, lowvram patches: 0
[INFO] loaded completely; 17375.81 MB usable, 14960.20 MB loaded, full load: True
[WARNING] Warning: TAESD previews enabled, but could not find models/vae_approx/None
[WARNING] Warning: TAESD previews enabled, but could not find models/vae_approx/None
[INFO] Requested to load MiniMaxH3
[INFO] loaded partially; 18867.63 MB usable, 18680.06 MB loaded, 1316.09 MB offloaded, 257.27 MB buffer reserved, lowvram patches: 0
100%|██████████████████████████████████████████████████████████████████████████████████| 20/20 [00:38<00:00, 1.95s/it]
[WARNING] Warning: TAESD previews enabled, but could not find models/vae_approx/None
[INFO] 0 models unloaded.
[INFO] Unloaded partially: 661.87 MB freed, 18018.19 MB remains loaded, 257.27 MB buffer reserved, lowvram patches: 0
[INFO] Requested to load MiniMaxH3AudioVAE
[INFO] loaded completely; 2466.63 MB usable, 577.08 MB loaded, full load: True
[INFO] Requested to load MiniMaxH3VideoVAE
[INFO] Unloaded partially: 4043.47 MB freed, 13974.72 MB remains loaded, 661.52 MB buffer reserved, lowvram patches: 0
[INFO] loaded completely; 6806.72 MB usable, 4966.19 MB loaded, full load: True
[Pixaroma] Save Mp4 [save] — writing 124 frames @ 24fps (480x864, crf=19, yuv420p, +audio) -> Video_00011.mp4
[Pixaroma] Save Mp4 — saved F:\ComfyUI-Easy-Install\ComfyUI\output\Video_00011.mp4
[INFO] Prompt executed in 77.15 seconds
[INFO] got prompt
[WARNING] Warning: TAESD previews enabled, but could not find models/vae_approx/None
[WARNING] Warning: TAESD previews enabled, but could not find models/vae_approx/None
[INFO] Requested to load MiniMaxH3
[INFO] loaded partially; 18862.97 MB usable, 18643.29 MB loaded, 1352.86 MB offloaded, 257.27 MB buffer reserved, lowvram patches: 0
100%|██████████████████████████████████████████████████████████████████████████████████| 20/20 [00:43<00:00, 2.16s/it]
[WARNING] Warning: TAESD previews enabled, but could not find models/vae_approx/None
[INFO] 0 models unloaded.
[INFO] Unloaded partially: 735.41 MB freed, 17907.88 MB remains loaded, 257.27 MB buffer reserved, lowvram patches: 0
[INFO] Requested to load MiniMaxH3AudioVAE
[INFO] loaded completely; 2398.95 MB usable, 577.08 MB loaded, full load: True
[INFO] Requested to load MiniMaxH3VideoVAE
[INFO] Unloaded partially: 4006.68 MB freed, 13901.20 MB remains loaded, 661.52 MB buffer reserved, lowvram patches: 0
[INFO] loaded completely; 6880.24 MB usable, 4966.19 MB loaded, full load: True
[Pixaroma] Save Mp4 [save] — writing 124 frames @ 24fps (480x864, crf=19, yuv420p, +audio) -> Video_00012.mp4
[Pixaroma] Save Mp4 — saved F:\ComfyUI-Easy-Install\ComfyUI\output\Video_00012.mp4
[INFO] Prompt executed in 71.84 seconds
[INFO] got prompt
[WARNING] Warning: TAESD previews enabled, but could not find models/vae_approx/None
[WARNING] Warning: TAESD previews enabled, but could not find models/vae_approx/None
[INFO] Requested to load MiniMaxH3
[INFO] loaded partially; 18862.97 MB usable, 18643.29 MB loaded, 1352.86 MB offloaded, 257.27 MB buffer reserved, lowvram patches: 0
100%|██████████████████████████████████████████████████████████████████████████████████| 20/20 [01:06<00:00, 3.32s/it]
[INFO] Requested to load MiniMaxH3AudioVAE
[INFO] loaded completely; 1787.34 MB usable, 577.08 MB loaded, full load: True
[INFO] Requested to load MiniMaxH3VideoVAE
[INFO] Unloaded partially: 4742.09 MB freed, 13901.20 MB remains loaded, 661.52 MB buffer reserved, lowvram patches: 0
[INFO] loaded completely; 6880.24 MB usable, 4966.19 MB loaded, full load: True
[Pixaroma] Save Mp4 [save] — writing 124 frames @ 24fps (480x864, crf=19, yuv420p, +audio) -> Video_00013.mp4
[Pixaroma] Save Mp4 — saved F:\ComfyUI-Easy-Install\ComfyUI\output\Video_00013.mp4
[INFO] Prompt executed in 80.83 seconds
[INFO] got prompt
[INFO] Requested to load MiniMaxH3TEModel_
[INFO] Unloaded partially: 1079.80 MB freed, 3886.69 MB remains loaded, 72.04 MB buffer reserved, lowvram patches: 0
[INFO] loaded completely; 17471.86 MB usable, 14960.20 MB loaded, full load: True
[INFO] loaded completely; 5564.51 MB usable, 4966.19 MB loaded, full load: True
[WARNING] Warning: TAESD previews enabled, but could not find models/vae_approx/None
[INFO] Requested to load MiniMaxH3
[INFO] loaded partially; 18906.12 MB usable, 18716.83 MB loaded, 1279.32 MB offloaded, 257.27 MB buffer reserved, lowvram patches: 0
5%|████▏ | 1/20 [00:04<01:25, 4.50s/it][INFO] Interrupting prompt 6c521dff-f802-49ff-aaab-eee7230c647b
5%|████▏ | 1/20 [00:08<02:45, 8.71s/it]
[INFO] Processing interrupted
[INFO] Prompt executed in 24.53 seconds
[INFO] got prompt
[INFO] Requested to load MiniMaxH3TEModel_
[INFO] Unloaded partially: 15333.31 MB freed, 3383.52 MB remains loaded, 918.83 MB buffer reserved, lowvram patches: 0
[INFO] loaded completely; 17973.03 MB usable, 14960.20 MB loaded, full load: True
[INFO] Requested to load MiniMaxH3VideoVAE
[INFO] Unloaded partially: 3383.52 MB freed, 0.00 MB remains loaded, 918.83 MB buffer reserved, lowvram patches: 0
[INFO] loaded completely; 6396.36 MB usable, 4966.19 MB loaded, full load: True
[WARNING] Warning: TAESD previews enabled, but could not find models/vae_approx/None
[INFO] loaded partially; 18906.12 MB usable, 18716.83 MB loaded, 1279.32 MB offloaded, 257.27 MB buffer reserved, lowvram patches: 0
10%|████████▎ | 2/20 [00:09<01:21, 4.50s/it][INFO] Interrupting prompt d2b32306-f682-4d55-b692-6d4a597c9eea
10%|████████▎ | 2/20 [00:10<01:36, 5.37s/it]
[INFO] Processing interrupted
[INFO] Prompt executed in 25.99 seconds
[INFO] got prompt
[WARNING] Warning: TAESD previews enabled, but could not find models/vae_approx/None
[INFO] 0 models unloaded.
[INFO] Unloaded partially: 735.41 MB freed, 17981.42 MB remains loaded, 257.27 MB buffer reserved, lowvram patches: 0
100%|██████████████████████████████████████████████████████████████████████████████████| 20/20 [01:27<00:00, 4.39s/it]
[INFO] Requested to load MiniMaxH3AudioVAE
[INFO] loaded completely; 2886.82 MB usable, 577.08 MB loaded, full load: True
[INFO] Requested to load MiniMaxH3VideoVAE
[INFO] Unloaded partially: 3786.13 MB freed, 14195.28 MB remains loaded, 661.52 MB buffer reserved, lowvram patches: 0
[INFO] loaded completely; 6584.16 MB usable, 4966.19 MB loaded, full load: True
[INFO] Prompt executed in 97.17 seconds
[INFO] got prompt
[INFO] Requested to load MiniMaxH3TEModel_
[INFO] Unloaded partially: 983.75 MB freed, 3982.74 MB remains loaded, 72.04 MB buffer reserved, lowvram patches: 0
[INFO] loaded completely; 17373.81 MB usable, 14960.20 MB loaded, full load: True
[INFO] loaded completely; 6032.04 MB usable, 4966.19 MB loaded, full load: True
[WARNING] Warning: TAESD previews enabled, but could not find models/vae_approx/None
[INFO] Requested to load MiniMaxH3
[INFO] loaded partially; 13982.20 MB usable, 13607.12 MB loaded, 6389.03 MB offloaded, 661.52 MB buffer reserved, lowvram patches: 0
50%|█████████████████████████████████████████ | 10/20 [03:16<03:20, 20.02s/it][INFO] Interrupting prompt d36e9314-fed9-4471-a7f7-36f57f7de3c6
50%|█████████████████████████████████████████ | 10/20 [03:28<03:28, 20.89s/it]
[INFO] Processing interrupted
[INFO] Prompt executed in 224.17 seconds
[INFO] got prompt
[INFO] Requested to load MiniMaxH3TEModel_
[INFO] Unloaded partially: 9046.73 MB freed, 4560.39 MB remains loaded, 918.83 MB buffer reserved, lowvram patches: 0
[INFO] loaded completely; 16796.16 MB usable, 14960.20 MB loaded, full load: True
[INFO] Requested to load MiniMaxH3VideoVAE
[INFO] Unloaded partially: 4266.17 MB freed, 294.22 MB remains loaded, 918.83 MB buffer reserved, lowvram patches: 0
[INFO] loaded completely; 6102.15 MB usable, 4966.19 MB loaded, full load: True
[WARNING] Warning: TAESD previews enabled, but could not find models/vae_approx/None
[INFO] loaded partially; 18792.74 MB usable, 18606.52 MB loaded, 1389.63 MB offloaded, 257.27 MB buffer reserved, lowvram patches: 0
25%|████████████████████▊ | 5/20 [00:23<01:10, 4.72s/it][INFO] Interrupting prompt 1a931c2d-5da1-479e-a281-1c28194010af
25%|████████████████████▊ | 5/20 [00:27<01:21, 5.45s/it]
[INFO] Processing interrupted
[INFO] Prompt executed in 42.09 seconds
[INFO] got prompt
[WARNING] Warning: TAESD previews enabled, but could not find models/vae_approx/None
[INFO] 0 models unloaded.
[INFO] Unloaded partially: 882.41 MB freed, 17724.11 MB remains loaded, 257.27 MB buffer reserved, lowvram patches: 0
100%|██████████████████████████████████████████████████████████████████████████████████| 20/20 [01:35<00:00, 4.79s/it]
[INFO] Requested to load MiniMaxH3AudioVAE
[INFO] loaded completely; 2955.56 MB usable, 577.08 MB loaded, full load: True
[INFO] Requested to load MiniMaxH3VideoVAE
[INFO] Unloaded partially: 3528.82 MB freed, 14195.28 MB remains loaded, 661.52 MB buffer reserved, lowvram patches: 0
[INFO] loaded completely; 6584.16 MB usable, 4966.19 MB loaded, full load: True
[INFO] Prompt executed in 107.23 seconds
```

### Other

Start with the official video_minimax_h3_i2v template and trace the named MiniMaxH3ImageToVideo and VAEDecode(video_vae_fp16) nodes at 1344x768. Reproduce with the listed models, then compare the int8_convrot and bf16 variants or isolate the template to identify whether the decode path is responsible; done means continuous frames without square-tile seams.

Written by the indexing model from the issue text.