LTX2.3 Windows fatal exception: access violation - after comfy update 0.17.0

[LL9Zarnek]

Custom Node Testing

• I have tried disabling custom nodes and the issue persists (see how to disable custom nodes if you need help)

Expected Behavior

video output successfully as before updating, the issue i don't even know what commit i was on that was working

Actual Behavior

Windows fatal exception: access violation

Steps to Reproduce

update comfy to latest 0.17.0 and use ltx2.3 i2v workflow at 1920x1088 length 241

Debug Logs

[START] Security scan
[DONE] Security scan
## ComfyUI-Manager: installing dependencies done.
** ComfyUI startup time: 2026-03-16 19:25:25.752
** 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: I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\python_embeded\python.exe
** ComfyUI Path: I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI
** ComfyUI Base Folder Path: I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI
** User directory: I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\user
** ComfyUI-Manager config path: I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\user\__manager\config.ini
** Log path: I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\user\comfyui.log

Prestartup times for custom nodes:
   0.0 seconds: I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\custom_nodes\rgthree-comfy
   0.0 seconds: I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-Easy-Use
   3.4 seconds: I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-manager

Found comfy_kitchen backend cuda: {'available': True, 'disabled': False, 'unavailable_reason': None, 'capabilities': ['apply_rope', 'apply_rope1', 'dequantize_nvfp4', 'dequantize_per_tensor_fp8', 'quantize_mxfp8', 'quantize_nvfp4', 'quantize_per_tensor_fp8', 'scaled_mm_nvfp4']}
Found comfy_kitchen backend triton: {'available': True, 'disabled': True, 'unavailable_reason': None, 'capabilities': ['apply_rope', 'apply_rope1', 'dequantize_nvfp4', 'dequantize_per_tensor_fp8', 'quantize_mxfp8', 'quantize_nvfp4', 'quantize_per_tensor_fp8']}
Found comfy_kitchen backend eager: {'available': True, 'disabled': False, 'unavailable_reason': None, 'capabilities': ['apply_rope', 'apply_rope1', 'dequantize_mxfp8', 'dequantize_nvfp4', 'dequantize_per_tensor_fp8', 'quantize_mxfp8', 'quantize_nvfp4', 'quantize_per_tensor_fp8', 'scaled_mm_mxfp8', 'scaled_mm_nvfp4']}
Checkpoint files will always be loaded safely.
Total VRAM 32606 MB, total RAM 32701 MB
pytorch version: 2.9.1+cu130
Enabled fp16 accumulation.
Set vram state to: NORMAL_VRAM
Disabling smart memory management
Device: cuda:0 NVIDIA GeForce RTX 5090 : cudaMallocAsync
Using async weight offloading with 2 streams
Enabled pinned memory 14715.0
working around nvidia conv3d memory bug.
Using sage attention
aimdo: src-win/cuda-detour.c:77:INFO:aimdo_setup_hooks: found driver at 00007FFD3DE10000, installing 4 hooks
aimdo: src-win/cuda-detour.c:61:DEBUG:install_hook_entrys: hooks successfully installed
aimdo: src/control.c:69:INFO:comfy-aimdo inited for GPU: NVIDIA GeForce RTX 5090 (VRAM: 32606 MB)
DynamicVRAM support detected and enabled
Python version: 3.12.10 (tags/v3.12.10:0cc8128, Apr  8 2025, 12:21:36) [MSC v.1943 64 bit (AMD64)]
ComfyUI version: 0.17.0
comfy-aimdo version: 0.2.12
comfy-kitchen version: 0.2.8
ComfyUI frontend version: 1.41.20
[Prompt Server] web root: I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\python_embeded\Lib\site-packages\comfyui_frontend_package\static
Asset seeder disabled
[ComfyUI-Easy-Use] server: v1.3.7 Loaded
[ComfyUI-Easy-Use] web root: I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-Easy-Use\web_version/v2 Loaded
ComfyUI-GGUF: Allowing full torch compile
### Loading: ComfyUI-Manager (V3.39.2)
[ComfyUI-Manager] network_mode: public
[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'.
### ComfyUI Version: v0.17.0-17-g593be209 | Released on '2026-03-15'
======================================== ComfyUI-nunchaku Initialization ========================================
Nunchaku version: 1.2.1
ComfyUI-nunchaku version: 1.2.1
[ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/alter-list.json
[ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/model-list.json
[ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/github-stats.json
[ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/extension-node-map.json
[ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/custom-node-list.json
=================================================================================================================
[ComfyUI-RMBG] v3.0.0 | 43 nodes Loaded
[I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui_controlnet_aux] | INFO -> Using ckpts path: I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui_controlnet_aux\ckpts
[I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui_controlnet_aux] | INFO -> Using symlinks: False
[I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui_controlnet_aux] | INFO -> Using ort providers: ['CUDAExecutionProvider', 'DirectMLExecutionProvider', 'OpenVINOExecutionProvider', 'ROCMExecutionProvider', 'CPUExecutionProvider', 'CoreMLExecutionProvider']
I:\Ai\ComfyUI_RTX5090\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")

Initializing ControlAltAI Nodes
✅ ComfyUI-Qwen-TTS v1.0.6 loaded

[rgthree-comfy] Loaded 48 extraordinary 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.

⚡ SeedVR2 optimizations check: SageAttention ✅ | Flash Attention ✅ | Triton ✅
🔧 Conv3d workaround active: PyTorch 2.9.1, cuDNN 91200 (fixing VAE 3x memory bug)
📊 Initial CUDA memory: 30.15GB free / 31.84GB total

Import times for custom nodes:
   0.0 seconds: I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\custom_nodes\websocket_image_save.py
   0.0 seconds: I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-inpaint-cropandstitch
   0.0 seconds: I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-SCAIL-Pose
   0.0 seconds: I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-GGUF
   0.0 seconds: I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-TiledDiffusion
   0.0 seconds: I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-WanAnimatePreprocess
   0.0 seconds: I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-MelBandRoFormer
   0.0 seconds: I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\custom_nodes\controlaltai-nodes
   0.0 seconds: I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-KJNodes
   0.0 seconds: I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\custom_nodes\rgthree-comfy
   0.1 seconds: I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-QwenVL
   0.1 seconds: I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui_controlnet_aux
   0.2 seconds: I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui_layerstyle
   0.2 seconds: I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\custom_nodes\seedvr2_videoupscaler
   0.4 seconds: I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\custom_nodes\qwen3-tts-comfyui
   0.4 seconds: I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-itools
   0.4 seconds: I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-rmbg
   0.4 seconds: I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-WanVideoWrapper
   0.5 seconds: I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-easy-sam3
   0.5 seconds: I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-manager
   1.6 seconds: I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\custom_nodes\comfyui-videohelpersuite
   1.7 seconds: I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-nunchaku
   5.4 seconds: I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-Easy-Use

Context impl SQLiteImpl.
Will assume non-transactional DDL.
Starting server

To see the GUI go to: http://127.0.0.1:8188
FETCH ComfyRegistry Data: 5/131
[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.
[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.
[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.
[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.
[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.
FETCH ComfyRegistry Data: 10/131
FETCH ComfyRegistry Data: 15/131
got prompt
Found quantization metadata version 1
Detected mixed precision quantization
Using mixed precision operations
model weight dtype torch.float16, manual cast: torch.bfloat16
model_type FLUX
VAE load device: cuda:0, offload device: cpu, dtype: torch.bfloat16
no CLIP/text encoder weights in checkpoint, the text encoder model will not be loaded.
Requested to load VideoVAE
Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.
FETCH ComfyRegistry Data: 20/131
clip missing: ['multi_modal_projector.mm_input_projection_weight', 'multi_modal_projector.mm_soft_emb_norm.weight', 'vision_model.embeddings.patch_embedding.weight', 'vision_model.embeddings.patch_embedding.bias', 'vision_model.encoder.layers.0.layer_norm1.weight', 'vision_model.encoder.layers.0.layer_norm1.bias', 'vision_model.encoder.layers.0.self_attn.q_proj.weight', 'vision_model.encoder.layers.0.self_attn.q_proj.bias', 'vision_model.encoder.layers.0.self_attn.k_proj.weight', 'vision_model.encoder.layers.0.self_attn.k_proj.bias', 'vision_model.encoder.layers.0.self_attn.v_proj.weight', 'vision_model.encoder.layers.0.self_attn.v_proj.bias', 'vision_model.encoder.layers.0.self_attn.out_proj.weight', 'vision_model.encoder.layers.0.self_attn.out_proj.bias', 'vision_model.encoder.layers.0.layer_norm2.weight', 'vision_model.encoder.layers.0.layer_norm2.bias', 'vision_model.encoder.layers.0.mlp.fc1.weight', 'vision_model.encoder.layers.0.mlp.fc1.bias', 'vision_model.encoder.layers.0.mlp.fc2.weight', 'vision_model.encoder.layers.0.mlp.fc2.bias', 'vision_model.encoder.layers.1.layer_norm1.weight', 'vision_model.encoder.layers.1.layer_norm1.bias', 'vision_model.encoder.layers.1.self_attn.q_proj.weight', 'vision_model.encoder.layers.1.self_attn.q_proj.bias', 'vision_model.encoder.layers.1.self_attn.k_proj.weight', 'vision_model.encoder.layers.1.self_attn.k_proj.bias', 'vision_model.encoder.layers.1.self_attn.v_proj.weight', 'vision_model.encoder.layers.1.self_attn.v_proj.bias', 'vision_model.encoder.layers.1.self_attn.out_proj.weight', 'vision_model.encoder.layers.1.self_attn.out_proj.bias', 'vision_model.encoder.layers.1.layer_norm2.weight', 'vision_model.encoder.layers.1.layer_norm2.bias', 'vision_model.encoder.layers.1.mlp.fc1.weight', 'vision_model.encoder.layers.1.mlp.fc1.bias', 'vision_model.encoder.layers.1.mlp.fc2.weight', 'vision_model.encoder.layers.1.mlp.fc2.bias', 'vision_model.encoder.layers.2.layer_norm1.weight', 'vision_model.encoder.layers.2.layer_norm1.bias', 'vision_model.encoder.layers.2.self_attn.q_proj.weight', 'vision_model.encoder.layers.2.self_attn.q_proj.bias', 'vision_model.encoder.layers.2.self_attn.k_proj.weight', 'vision_model.encoder.layers.2.self_attn.k_proj.bias', 'vision_model.encoder.layers.2.self_attn.v_proj.weight', 'vision_model.encoder.layers.2.self_attn.v_proj.bias', 'vision_model.encoder.layers.2.self_attn.out_proj.weight', 'vision_model.encoder.layers.2.self_attn.out_proj.bias', 'vision_model.encoder.layers.2.layer_norm2.weight', 'vision_model.encoder.layers.2.layer_norm2.bias', 'vision_model.encoder.layers.2.mlp.fc1.weight', 'vision_model.encoder.layers.2.mlp.fc1.bias', 'vision_model.encoder.layers.2.mlp.fc2.weight', 'vision_model.encoder.layers.2.mlp.fc2.bias', 'vision_model.encoder.layers.3.layer_norm1.weight', 'vision_model.encoder.layers.3.layer_norm1.bias', 'vision_model.encoder.layers.3.self_attn.q_proj.weight', 'vision_model.encoder.layers.3.self_attn.q_proj.bias', 'vision_model.encoder.layers.3.self_attn.k_proj.weight', 'vision_model.encoder.layers.3.self_attn.k_proj.bias', 'vision_model.encoder.layers.3.self_attn.v_proj.weight', 'vision_model.encoder.layers.3.self_attn.v_proj.bias', 'vision_model.encoder.layers.3.self_attn.out_proj.weight', 'vision_model.encoder.layers.3.self_attn.out_proj.bias', 'vision_model.encoder.layers.3.layer_norm2.weight', 'vision_model.encoder.layers.3.layer_norm2.bias', 'vision_model.encoder.layers.3.mlp.fc1.weight', 'vision_model.encoder.layers.3.mlp.fc1.bias', 'vision_model.encoder.layers.3.mlp.fc2.weight', 'vision_model.encoder.layers.3.mlp.fc2.bias', 'vision_model.encoder.layers.4.layer_norm1.weight', 'vision_model.encoder.layers.4.layer_norm1.bias', 'vision_model.encoder.layers.4.self_attn.q_proj.weight', 'vision_model.encoder.layers.4.self_attn.q_proj.bias', 'vision_model.encoder.layers.4.self_attn.k_proj.weight', 'vision_model.encoder.layers.4.self_attn.k_proj.bias', 'vision_model.encoder.layers.4.self_attn.v_proj.weight', 'vision_model.encoder.layers.4.self_attn.v_proj.bias', 'vision_model.encoder.layers.4.self_attn.out_proj.weight', 'vision_model.encoder.layers.4.self_attn.out_proj.bias', 'vision_model.encoder.layers.4.layer_norm2.weight', 'vision_model.encoder.layers.4.layer_norm2.bias', 'vision_model.encoder.layers.4.mlp.fc1.weight', 'vision_model.encoder.layers.4.mlp.fc1.bias', 'vision_model.encoder.layers.4.mlp.fc2.weight', 'vision_model.encoder.layers.4.mlp.fc2.bias', 'vision_model.encoder.layers.5.layer_norm1.weight', 'vision_model.encoder.layers.5.layer_norm1.bias', 'vision_model.encoder.layers.5.self_attn.q_proj.weight', 'vision_model.encoder.layers.5.self_attn.q_proj.bias', 'vision_model.encoder.layers.5.self_attn.k_proj.weight', 'vision_model.encoder.layers.5.self_attn.k_proj.bias', 'vision_model.encoder.layers.5.self_attn.v_proj.weight', 'vision_model.encoder.layers.5.self_attn.v_proj.bias', 'vision_model.encoder.layers.5.self_attn.out_proj.weight', 'vision_model.encoder.layers.5.self_attn.out_proj.bias', 'vision_model.encoder.layers.5.layer_norm2.weight', 'vision_model.encoder.layers.5.layer_norm2.bias', 'vision_model.encoder.layers.5.mlp.fc1.weight', 'vision_model.encoder.layers.5.mlp.fc1.bias', 'vision_model.encoder.layers.5.mlp.fc2.weight', 'vision_model.encoder.layers.5.mlp.fc2.bias', 'vision_model.encoder.layers.6.layer_norm1.weight', 'vision_model.encoder.layers.6.layer_norm1.bias', 'vision_model.encoder.layers.6.self_attn.q_proj.weight', 'vision_model.encoder.layers.6.self_attn.q_proj.bias', 'vision_model.encoder.layers.6.self_attn.k_proj.weight', 'vision_model.encoder.layers.6.self_attn.k_proj.bias', 'vision_model.encoder.layers.6.self_attn.v_proj.weight', 'vision_model.encoder.layers.6.self_attn.v_proj.bias', 'vision_model.encoder.layers.6.self_attn.out_proj.weight', 'vision_model.encoder.layers.6.self_attn.out_proj.bias', 'vision_model.encoder.layers.6.layer_norm2.weight', 'vision_model.encoder.layers.6.layer_norm2.bias', 'vision_model.encoder.layers.6.mlp.fc1.weight', 'vision_model.encoder.layers.6.mlp.fc1.bias', 'vision_model.encoder.layers.6.mlp.fc2.weight', 'vision_model.encoder.layers.6.mlp.fc2.bias', 'vision_model.encoder.layers.7.layer_norm1.weight', 'vision_model.encoder.layers.7.layer_norm1.bias', 'vision_model.encoder.layers.7.self_attn.q_proj.weight', 'vision_model.encoder.layers.7.self_attn.q_proj.bias', 'vision_model.encoder.layers.7.self_attn.k_proj.weight', 'vision_model.encoder.layers.7.self_attn.k_proj.bias', 'vision_model.encoder.layers.7.self_attn.v_proj.weight', 'vision_model.encoder.layers.7.self_attn.v_proj.bias', 'vision_model.encoder.layers.7.self_attn.out_proj.weight', 'vision_model.encoder.layers.7.self_attn.out_proj.bias', 'vision_model.encoder.layers.7.layer_norm2.weight', 'vision_model.encoder.layers.7.layer_norm2.bias', 'vision_model.encoder.layers.7.mlp.fc1.weight', 'vision_model.encoder.layers.7.mlp.fc1.bias', 'vision_model.encoder.layers.7.mlp.fc2.weight', 'vision_model.encoder.layers.7.mlp.fc2.bias', 'vision_model.encoder.layers.8.layer_norm1.weight', 'vision_model.encoder.layers.8.layer_norm1.bias', 'vision_model.encoder.layers.8.self_attn.q_proj.weight', 'vision_model.encoder.layers.8.self_attn.q_proj.bias', 'vision_model.encoder.layers.8.self_attn.k_proj.weight', 'vision_model.encoder.layers.8.self_attn.k_proj.bias', 'vision_model.encoder.layers.8.self_attn.v_proj.weight', 'vision_model.encoder.layers.8.self_attn.v_proj.bias', 'vision_model.encoder.layers.8.self_attn.out_proj.weight', 'vision_model.encoder.layers.8.self_attn.out_proj.bias', 'vision_model.encoder.layers.8.layer_norm2.weight', 'vision_model.encoder.layers.8.layer_norm2.bias', 'vision_model.encoder.layers.8.mlp.fc1.weight', 'vision_model.encoder.layers.8.mlp.fc1.bias', 'vision_model.encoder.layers.8.mlp.fc2.weight', 'vision_model.encoder.layers.8.mlp.fc2.bias', 'vision_model.encoder.layers.9.layer_norm1.weight', 'vision_model.encoder.layers.9.layer_norm1.bias', 'vision_model.encoder.layers.9.self_attn.q_proj.weight', 'vision_model.encoder.layers.9.self_attn.q_proj.bias', 'vision_model.encoder.layers.9.self_attn.k_proj.weight', 'vision_model.encoder.layers.9.self_attn.k_proj.bias', 'vision_model.encoder.layers.9.self_attn.v_proj.weight', 'vision_model.encoder.layers.9.self_attn.v_proj.bias', 'vision_model.encoder.layers.9.self_attn.out_proj.weight', 'vision_model.encoder.layers.9.self_attn.out_proj.bias', 'vision_model.encoder.layers.9.layer_norm2.weight', 'vision_model.encoder.layers.9.layer_norm2.bias', 'vision_model.encoder.layers.9.mlp.fc1.weight', 'vision_model.encoder.layers.9.mlp.fc1.bias', 'vision_model.encoder.layers.9.mlp.fc2.weight', 'vision_model.encoder.layers.9.mlp.fc2.bias', 'vision_model.encoder.layers.10.layer_norm1.weight', 'vision_model.encoder.layers.10.layer_norm1.bias', 'vision_model.encoder.layers.10.self_attn.q_proj.weight', 'vision_model.encoder.layers.10.self_attn.q_proj.bias', 'vision_model.encoder.layers.10.self_attn.k_proj.weight', 'vision_model.encoder.layers.10.self_attn.k_proj.bias', 'vision_model.encoder.layers.10.self_attn.v_proj.weight', 'vision_model.encoder.layers.10.self_attn.v_proj.bias', 'vision_model.encoder.layers.10.self_attn.out_proj.weight', 'vision_model.encoder.layers.10.self_attn.out_proj.bias', 'vision_model.encoder.layers.10.layer_norm2.weight', 'vision_model.encoder.layers.10.layer_norm2.bias', 'vision_model.encoder.layers.10.mlp.fc1.weight', 'vision_model.encoder.layers.10.mlp.fc1.bias', 'vision_model.encoder.layers.10.mlp.fc2.weight', 'vision_model.encoder.layers.10.mlp.fc2.bias', 'vision_model.encoder.layers.11.layer_norm1.weight', 'vision_model.encoder.layers.11.layer_norm1.bias', 'vision_model.encoder.layers.11.self_attn.q_proj.weight', 'vision_model.encoder.layers.11.self_attn.q_proj.bias', 'vision_model.encoder.layers.11.self_attn.k_proj.weight', 'vision_model.encoder.layers.11.self_attn.k_proj.bias', 'vision_model.encoder.layers.11.self_attn.v_proj.weight', 'vision_model.encoder.layers.11.self_attn.v_proj.bias', 'vision_model.encoder.layers.11.self_attn.out_proj.weight', 'vision_model.encoder.layers.11.self_attn.out_proj.bias', 'vision_model.encoder.layers.11.layer_norm2.weight', 'vision_model.encoder.layers.11.layer_norm2.bias', 'vision_model.encoder.layers.11.mlp.fc1.weight', 'vision_model.encoder.layers.11.mlp.fc1.bias', 'vision_model.encoder.layers.11.mlp.fc2.weight', 'vision_model.encoder.layers.11.mlp.fc2.bias', 'vision_model.encoder.layers.12.layer_norm1.weight', 'vision_model.encoder.layers.12.layer_norm1.bias', 'vision_model.encoder.layers.12.self_attn.q_proj.weight', 'vision_model.encoder.layers.12.self_attn.q_proj.bias', 'vision_model.encoder.layers.12.self_attn.k_proj.weight', 'vision_model.encoder.layers.12.self_attn.k_proj.bias', 'vision_model.encoder.layers.12.self_attn.v_proj.weight', 'vision_model.encoder.layers.12.self_attn.v_proj.bias', 'vision_model.encoder.layers.12.self_attn.out_proj.weight', 'vision_model.encoder.layers.12.self_attn.out_proj.bias', 'vision_model.encoder.layers.12.layer_norm2.weight', 'vision_model.encoder.layers.12.layer_norm2.bias', 'vision_model.encoder.layers.12.mlp.fc1.weight', 'vision_model.encoder.layers.12.mlp.fc1.bias', 'vision_model.encoder.layers.12.mlp.fc2.weight', 'vision_model.encoder.layers.12.mlp.fc2.bias', 'vision_model.encoder.layers.13.layer_norm1.weight', 'vision_model.encoder.layers.13.layer_norm1.bias', 'vision_model.encoder.layers.13.self_attn.q_proj.weight', 'vision_model.encoder.layers.13.self_attn.q_proj.bias', 'vision_model.encoder.layers.13.self_attn.k_proj.weight', 'vision_model.encoder.layers.13.self_attn.k_proj.bias', 'vision_model.encoder.layers.13.self_attn.v_proj.weight', 'vision_model.encoder.layers.13.self_attn.v_proj.bias', 'vision_model.encoder.layers.13.self_attn.out_proj.weight', 'vision_model.encoder.layers.13.self_attn.out_proj.bias', 'vision_model.encoder.layers.13.layer_norm2.weight', 'vision_model.encoder.layers.13.layer_norm2.bias', 'vision_model.encoder.layers.13.mlp.fc1.weight', 'vision_model.encoder.layers.13.mlp.fc1.bias', 'vision_model.encoder.layers.13.mlp.fc2.weight', 'vision_model.encoder.layers.13.mlp.fc2.bias', 'vision_model.encoder.layers.14.layer_norm1.weight', 'vision_model.encoder.layers.14.layer_norm1.bias', 'vision_model.encoder.layers.14.self_attn.q_proj.weight', 'vision_model.encoder.layers.14.self_attn.q_proj.bias', 'vision_model.encoder.layers.14.self_attn.k_proj.weight', 'vision_model.encoder.layers.14.self_attn.k_proj.bias', 'vision_model.encoder.layers.14.self_attn.v_proj.weight', 'vision_model.encoder.layers.14.self_attn.v_proj.bias', 'vision_model.encoder.layers.14.self_attn.out_proj.weight', 'vision_model.encoder.layers.14.self_attn.out_proj.bias', 'vision_model.encoder.layers.14.layer_norm2.weight', 'vision_model.encoder.layers.14.layer_norm2.bias', 'vision_model.encoder.layers.14.mlp.fc1.weight', 'vision_model.encoder.layers.14.mlp.fc1.bias', 'vision_model.encoder.layers.14.mlp.fc2.weight', 'vision_model.encoder.layers.14.mlp.fc2.bias', 'vision_model.encoder.layers.15.layer_norm1.weight', 'vision_model.encoder.layers.15.layer_norm1.bias', 'vision_model.encoder.layers.15.self_attn.q_proj.weight', 'vision_model.encoder.layers.15.self_attn.q_proj.bias', 'vision_model.encoder.layers.15.self_attn.k_proj.weight', 'vision_model.encoder.layers.15.self_attn.k_proj.bias', 'vision_model.encoder.layers.15.self_attn.v_proj.weight', 'vision_model.encoder.layers.15.self_attn.v_proj.bias', 'vision_model.encoder.layers.15.self_attn.out_proj.weight', 'vision_model.encoder.layers.15.self_attn.out_proj.bias', 'vision_model.encoder.layers.15.layer_norm2.weight', 'vision_model.encoder.layers.15.layer_norm2.bias', 'vision_model.encoder.layers.15.mlp.fc1.weight', 'vision_model.encoder.layers.15.mlp.fc1.bias', 'vision_model.encoder.layers.15.mlp.fc2.weight', 'vision_model.encoder.layers.15.mlp.fc2.bias', 'vision_model.encoder.layers.16.layer_norm1.weight', 'vision_model.encoder.layers.16.layer_norm1.bias', 'vision_model.encoder.layers.16.self_attn.q_proj.weight', 'vision_model.encoder.layers.16.self_attn.q_proj.bias', 'vision_model.encoder.layers.16.self_attn.k_proj.weight', 'vision_model.encoder.layers.16.self_attn.k_proj.bias', 'vision_model.encoder.layers.16.self_attn.v_proj.weight', 'vision_model.encoder.layers.16.self_attn.v_proj.bias', 'vision_model.encoder.layers.16.self_attn.out_proj.weight', 'vision_model.encoder.layers.16.self_attn.out_proj.bias', 'vision_model.encoder.layers.16.layer_norm2.weight', 'vision_model.encoder.layers.16.layer_norm2.bias', 'vision_model.encoder.layers.16.mlp.fc1.weight', 'vision_model.encoder.layers.16.mlp.fc1.bias', 'vision_model.encoder.layers.16.mlp.fc2.weight', 'vision_model.encoder.layers.16.mlp.fc2.bias', 'vision_model.encoder.layers.17.layer_norm1.weight', 'vision_model.encoder.layers.17.layer_norm1.bias', 'vision_model.encoder.layers.17.self_attn.q_proj.weight', 'vision_model.encoder.layers.17.self_attn.q_proj.bias', 'vision_model.encoder.layers.17.self_attn.k_proj.weight', 'vision_model.encoder.layers.17.self_attn.k_proj.bias', 'vision_model.encoder.layers.17.self_attn.v_proj.weight', 'vision_model.encoder.layers.17.self_attn.v_proj.bias', 'vision_model.encoder.layers.17.self_attn.out_proj.weight', 'vision_model.encoder.layers.17.self_attn.out_proj.bias', 'vision_model.encoder.layers.17.layer_norm2.weight', 'vision_model.encoder.layers.17.layer_norm2.bias', 'vision_model.encoder.layers.17.mlp.fc1.weight', 'vision_model.encoder.layers.17.mlp.fc1.bias', 'vision_model.encoder.layers.17.mlp.fc2.weight', 'vision_model.encoder.layers.17.mlp.fc2.bias', 'vision_model.encoder.layers.18.layer_norm1.weight', 'vision_model.encoder.layers.18.layer_norm1.bias', 'vision_model.encoder.layers.18.self_attn.q_proj.weight', 'vision_model.encoder.layers.18.self_attn.q_proj.bias', 'vision_model.encoder.layers.18.self_attn.k_proj.weight', 'vision_model.encoder.layers.18.self_attn.k_proj.bias', 'vision_model.encoder.layers.18.self_attn.v_proj.weight', 'vision_model.encoder.layers.18.self_attn.v_proj.bias', 'vision_model.encoder.layers.18.self_attn.out_proj.weight', 'vision_model.encoder.layers.18.self_attn.out_proj.bias', 'vision_model.encoder.layers.18.layer_norm2.weight', 'vision_model.encoder.layers.18.layer_norm2.bias', 'vision_model.encoder.layers.18.mlp.fc1.weight', 'vision_model.encoder.layers.18.mlp.fc1.bias', 'vision_model.encoder.layers.18.mlp.fc2.weight', 'vision_model.encoder.layers.18.mlp.fc2.bias', 'vision_model.encoder.layers.19.layer_norm1.weight', 'vision_model.encoder.layers.19.layer_norm1.bias', 'vision_model.encoder.layers.19.self_attn.q_proj.weight', 'vision_model.encoder.layers.19.self_attn.q_proj.bias', 'vision_model.encoder.layers.19.self_attn.k_proj.weight', 'vision_model.encoder.layers.19.self_attn.k_proj.bias', 'vision_model.encoder.layers.19.self_attn.v_proj.weight', 'vision_model.encoder.layers.19.self_attn.v_proj.bias', 'vision_model.encoder.layers.19.self_attn.out_proj.weight', 'vision_model.encoder.layers.19.self_attn.out_proj.bias', 'vision_model.encoder.layers.19.layer_norm2.weight', 'vision_model.encoder.layers.19.layer_norm2.bias', 'vision_model.encoder.layers.19.mlp.fc1.weight', 'vision_model.encoder.layers.19.mlp.fc1.bias', 'vision_model.encoder.layers.19.mlp.fc2.weight', 'vision_model.encoder.layers.19.mlp.fc2.bias', 'vision_model.encoder.layers.20.layer_norm1.weight', 'vision_model.encoder.layers.20.layer_norm1.bias', 'vision_model.encoder.layers.20.self_attn.q_proj.weight', 'vision_model.encoder.layers.20.self_attn.q_proj.bias', 'vision_model.encoder.layers.20.self_attn.k_proj.weight', 'vision_model.encoder.layers.20.self_attn.k_proj.bias', 'vision_model.encoder.layers.20.self_attn.v_proj.weight', 'vision_model.encoder.layers.20.self_attn.v_proj.bias', 'vision_model.encoder.layers.20.self_attn.out_proj.weight', 'vision_model.encoder.layers.20.self_attn.out_proj.bias', 'vision_model.encoder.layers.20.layer_norm2.weight', 'vision_model.encoder.layers.20.layer_norm2.bias', 'vision_model.encoder.layers.20.mlp.fc1.weight', 'vision_model.encoder.layers.20.mlp.fc1.bias', 'vision_model.encoder.layers.20.mlp.fc2.weight', 'vision_model.encoder.layers.20.mlp.fc2.bias', 'vision_model.encoder.layers.21.layer_norm1.weight', 'vision_model.encoder.layers.21.layer_norm1.bias', 'vision_model.encoder.layers.21.self_attn.q_proj.weight', 'vision_model.encoder.layers.21.self_attn.q_proj.bias', 'vision_model.encoder.layers.21.self_attn.k_proj.weight', 'vision_model.encoder.layers.21.self_attn.k_proj.bias', 'vision_model.encoder.layers.21.self_attn.v_proj.weight', 'vision_model.encoder.layers.21.self_attn.v_proj.bias', 'vision_model.encoder.layers.21.self_attn.out_proj.weight', 'vision_model.encoder.layers.21.self_attn.out_proj.bias', 'vision_model.encoder.layers.21.layer_norm2.weight', 'vision_model.encoder.layers.21.layer_norm2.bias', 'vision_model.encoder.layers.21.mlp.fc1.weight', 'vision_model.encoder.layers.21.mlp.fc1.bias', 'vision_model.encoder.layers.21.mlp.fc2.weight', 'vision_model.encoder.layers.21.mlp.fc2.bias', 'vision_model.encoder.layers.22.layer_norm1.weight', 'vision_model.encoder.layers.22.layer_norm1.bias', 'vision_model.encoder.layers.22.self_attn.q_proj.weight', 'vision_model.encoder.layers.22.self_attn.q_proj.bias', 'vision_model.encoder.layers.22.self_attn.k_proj.weight', 'vision_model.encoder.layers.22.self_attn.k_proj.bias', 'vision_model.encoder.layers.22.self_attn.v_proj.weight', 'vision_model.encoder.layers.22.self_attn.v_proj.bias', 'vision_model.encoder.layers.22.self_attn.out_proj.weight', 'vision_model.encoder.layers.22.self_attn.out_proj.bias', 'vision_model.encoder.layers.22.layer_norm2.weight', 'vision_model.encoder.layers.22.layer_norm2.bias', 'vision_model.encoder.layers.22.mlp.fc1.weight', 'vision_model.encoder.layers.22.mlp.fc1.bias', 'vision_model.encoder.layers.22.mlp.fc2.weight', 'vision_model.encoder.layers.22.mlp.fc2.bias', 'vision_model.encoder.layers.23.layer_norm1.weight', 'vision_model.encoder.layers.23.layer_norm1.bias', 'vision_model.encoder.layers.23.self_attn.q_proj.weight', 'vision_model.encoder.layers.23.self_attn.q_proj.bias', 'vision_model.encoder.layers.23.self_attn.k_proj.weight', 'vision_model.encoder.layers.23.self_attn.k_proj.bias', 'vision_model.encoder.layers.23.self_attn.v_proj.weight', 'vision_model.encoder.layers.23.self_attn.v_proj.bias', 'vision_model.encoder.layers.23.self_attn.out_proj.weight', 'vision_model.encoder.layers.23.self_attn.out_proj.bias', 'vision_model.encoder.layers.23.layer_norm2.weight', 'vision_model.encoder.layers.23.layer_norm2.bias', 'vision_model.encoder.layers.23.mlp.fc1.weight', 'vision_model.encoder.layers.23.mlp.fc1.bias', 'vision_model.encoder.layers.23.mlp.fc2.weight', 'vision_model.encoder.layers.23.mlp.fc2.bias', 'vision_model.encoder.layers.24.layer_norm1.weight', 'vision_model.encoder.layers.24.layer_norm1.bias', 'vision_model.encoder.layers.24.self_attn.q_proj.weight', 'vision_model.encoder.layers.24.self_attn.q_proj.bias', 'vision_model.encoder.layers.24.self_attn.k_proj.weight', 'vision_model.encoder.layers.24.self_attn.k_proj.bias', 'vision_model.encoder.layers.24.self_attn.v_proj.weight', 'vision_model.encoder.layers.24.self_attn.v_proj.bias', 'vision_model.encoder.layers.24.self_attn.out_proj.weight', 'vision_model.encoder.layers.24.self_attn.out_proj.bias', 'vision_model.encoder.layers.24.layer_norm2.weight', 'vision_model.encoder.layers.24.layer_norm2.bias', 'vision_model.encoder.layers.24.mlp.fc1.weight', 'vision_model.encoder.layers.24.mlp.fc1.bias', 'vision_model.encoder.layers.24.mlp.fc2.weight', 'vision_model.encoder.layers.24.mlp.fc2.bias', 'vision_model.encoder.layers.25.layer_norm1.weight', 'vision_model.encoder.layers.25.layer_norm1.bias', 'vision_model.encoder.layers.25.self_attn.q_proj.weight', 'vision_model.encoder.layers.25.self_attn.q_proj.bias', 'vision_model.encoder.layers.25.self_attn.k_proj.weight', 'vision_model.encoder.layers.25.self_attn.k_proj.bias', 'vision_model.encoder.layers.25.self_attn.v_proj.weight', 'vision_model.encoder.layers.25.self_attn.v_proj.bias', 'vision_model.encoder.layers.25.self_attn.out_proj.weight', 'vision_model.encoder.layers.25.self_attn.out_proj.bias', 'vision_model.encoder.layers.25.layer_norm2.weight', 'vision_model.encoder.layers.25.layer_norm2.bias', 'vision_model.encoder.layers.25.mlp.fc1.weight', 'vision_model.encoder.layers.25.mlp.fc1.bias', 'vision_model.encoder.layers.25.mlp.fc2.weight', 'vision_model.encoder.layers.25.mlp.fc2.bias', 'vision_model.encoder.layers.26.layer_norm1.weight', 'vision_model.encoder.layers.26.layer_norm1.bias', 'vision_model.encoder.layers.26.self_attn.q_proj.weight', 'vision_model.encoder.layers.26.self_attn.q_proj.bias', 'vision_model.encoder.layers.26.self_attn.k_proj.weight', 'vision_model.encoder.layers.26.self_attn.k_proj.bias', 'vision_model.encoder.layers.26.self_attn.v_proj.weight', 'vision_model.encoder.layers.26.self_attn.v_proj.bias', 'vision_model.encoder.layers.26.self_attn.out_proj.weight', 'vision_model.encoder.layers.26.self_attn.out_proj.bias', 'vision_model.encoder.layers.26.layer_norm2.weight', 'vision_model.encoder.layers.26.layer_norm2.bias', 'vision_model.encoder.layers.26.mlp.fc1.weight', 'vision_model.encoder.layers.26.mlp.fc1.bias', 'vision_model.encoder.layers.26.mlp.fc2.weight', 'vision_model.encoder.layers.26.mlp.fc2.bias', 'vision_model.post_layernorm.weight', 'vision_model.post_layernorm.bias']
CLIP/text encoder model load device: cuda:0, offload device: cpu, current: cpu, dtype: torch.float16
Requested to load LTXAVTEModel_
Model LTXAVTEModel_ prepared for dynamic VRAM loading. 25440MB Staged. 0 patches attached. Force pre-loaded 290 weights: 1497 KB.
FETCH ComfyRegistry Data: 25/131
FETCH ComfyRegistry Data: 30/131
Model LTXAVTEModel_ prepared for dynamic VRAM loading. 25440MB Staged. 0 patches attached. Force pre-loaded 290 weights: 1497 KB.
FETCH ComfyRegistry Data: 35/131
Requested to load LTXAV
Model LTXAV prepared for dynamic VRAM loading. 23838MB Staged. 1660 patches attached.
FETCH ComfyRegistry Data: 40/131
  0%|                                                                            | 0/8 [00:00<?, ?it/s,   Model Initializing ...  ]FETCH ComfyRegistry Data: 45/131
FETCH ComfyRegistry Data: 50/131
FETCH ComfyRegistry Data: 55/131
FETCH ComfyRegistry Data: 60/131
FETCH ComfyRegistry Data: 65/131
FETCH ComfyRegistry Data: 70/131
FETCH ComfyRegistry Data: 75/131
FETCH ComfyRegistry Data: 80/131
FETCH ComfyRegistry Data: 85/131
FETCH ComfyRegistry Data: 90/131
FETCH ComfyRegistry Data: 95/131
FETCH ComfyRegistry Data: 100/131
 25%|████████████████████████                                                                        | 2/8 [00:09<00:27,  4.58s/it]FETCH ComfyRegistry Data: 105/131
FETCH ComfyRegistry Data: 110/131
 38%|████████████████████████████████████                                                            | 3/8 [00:17<00:40,  8.14s/it]FETCH ComfyRegistry Data: 115/131
 50%|████████████████████████████████████████████████                                                | 4/8 [00:22<00:21,  5.43s/it]FETCH ComfyRegistry Data: 120/131
 62%|████████████████████████████████████████████████████████████                                    | 5/8 [00:28<00:16,  5.43s/it]FETCH ComfyRegistry Data: 125/131
 88%|████████████████████████████████████████████████████████████████████████████████████            | 7/8 [00:39<00:05,  5.45s/it]FETCH ComfyRegistry Data: 130/131
FETCH ComfyRegistry Data [DONE]
[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]
[ComfyUI-Manager] All startup tasks have been completed.
100%|████████████████████████████████████████████████████████████████████████████████████████████████| 8/8 [00:44<00:00,  5.62s/it]
Requested to load VideoVAE
Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.
Requested to load LTXAV
Model LTXAV prepared for dynamic VRAM loading. 23838MB Staged. 1660 patches attached.
100%|████████████████████████████████████████████████████████████████████████████████████████████████| 3/3 [01:38<00:00, 32.90s/it]
Requested to load AudioVAE
loaded completely; 28630.57 MB usable, 693.46 MB loaded, full load: True
Requested to load VideoVAE
0 models unloaded.
Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.
Windows fatal exception: access violation

Stack (most recent call first):
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\comfy\ldm\lightricks\vae\causal_video_autoencoder.py", line 564 in forward_orig
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\comfy\ldm\lightricks\vae\causal_video_autoencoder.py", line 572 in forward
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1786 in _call_impl
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1775 in _wrapped_call_impl
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\comfy\ldm\lightricks\vae\causal_video_autoencoder.py", line 1205 in decode
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\comfy\sd.py", line 902 in <lambda>
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\comfy\utils.py", line 1149 in tiled_scale_multidim
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\python_embeded\Lib\site-packages\torch\utils\_contextlib.py", line 120 in decorate_context
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\comfy\sd.py", line 903 in decode_tiled_3d
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\comfy\sd.py", line 1009 in decode_tiled
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\nodes.py", line 349 in decode
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\execution.py", line 296 in process_inputs
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\execution.py", line 308 in _async_map_node_over_list
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\execution.py", line 334 in get_output_data
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\execution.py", line 525 in execute
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\execution.py", line 758 in execute_async
  File "asyncio\events.py", line 88 in _run
  File "asyncio\base_events.py", line 1999 in _run_once
  File "asyncio\base_events.py", line 645 in run_forever
  File "asyncio\windows_events.py", line 322 in run_forever
  File "asyncio\base_events.py", line 678 in run_until_complete
  File "asyncio\runners.py", line 118 in run
  File "asyncio\runners.py", line 195 in run
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\execution.py", line 702 in execute
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\main.py", line 276 in prompt_worker
  File "threading.py", line 1012 in run
  File "threading.py", line 1075 in _bootstrap_inner
  File "threading.py", line 1032 in _bootstrap
Press any key to continue . . .

Other

No response

Fixed look below

[LL9Zarnek]

Trying potential fix commit

| comfyanonymous
Add --enable-dynamic-vram options to force enable it. (#13002)

7a16e8a · 2 hours ago
comfy
---
main.py
---

[ComfyUI-Easy-Use] server: v1.3.7 Loaded
[ComfyUI-Easy-Use] web root: I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\custom_nodes\ComfyUI-Easy-Use\web_version/v2 Loaded
ComfyUI-GGUF: Allowing full torch compile
### Loading: ComfyUI-Impact-Pack (V8.28.2)
[Impact Pack] Wildcard total size (0.00 MB) is within cache limit (50.00 MB). Using full cache mode.
[Impact Pack] Wildcards loading done.
### Loading: ComfyUI-Manager (V3.39.2)
[ComfyUI-Manager] network_mode: public
[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'.
### ComfyUI Version: v0.17.0-17-g593be209 | Released on '2026-03-15'

got prompt
Found quantization metadata version 1
Detected mixed precision quantization
Using mixed precision operations
model weight dtype torch.float16, manual cast: torch.bfloat16
model_type FLUX
FETCH ComfyRegistry Data: 90/131
VAE load device: cuda:0, offload device: cpu, dtype: torch.bfloat16
no CLIP/text encoder weights in checkpoint, the text encoder model will not be loaded.
Requested to load VideoVAE
Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.
  
CLIP/text encoder model load device: cuda:0, offload device: cpu, current: cpu, dtype: torch.float16
Requested to load LTXAVTEModel_
Model LTXAVTEModel_ prepared for dynamic VRAM loading. 25440MB Staged. 0 patches attached. Force pre-loaded 290 weights: 1497 KB.
FETCH ComfyRegistry Data: 100/131
FETCH ComfyRegistry Data: 105/131
Model LTXAVTEModel_ prepared for dynamic VRAM loading. 25440MB Staged. 0 patches attached. Force pre-loaded 290 weights: 1497 KB.
FETCH ComfyRegistry Data: 110/131
Requested to load LTXAV
Model LTXAV prepared for dynamic VRAM loading. 23838MB Staged. 1660 patches attached.
FETCH ComfyRegistry Data: 115/131
  0%|                                                                            | 0/8 [00:00<?, ?it/s,   Model Initializing ...  ]FETCH ComfyRegistry Data: 120/131
FETCH ComfyRegistry Data: 125/131
FETCH ComfyRegistry Data: 130/131
FETCH ComfyRegistry Data [DONE]
[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]
[ComfyUI-Manager] All startup tasks have been completed.
100%|████████████████████████████████████████████████████████████████████████████████████████████████| 8/8 [00:09<00:00,  1.15s/it]
Requested to load VideoVAE
Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.
Requested to load AudioVAE
loaded completely; 28630.57 MB usable, 693.46 MB loaded, full load: True
!!! Exception during processing !!! too many values to unpack (expected 4)
Traceback (most recent call last):
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\execution.py", line 525, in execute
    output_data, output_ui, has_subgraph, has_pending_tasks = await get_output_data(prompt_id, unique_id, obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, v3_data=v3_data)
                                                              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\execution.py", line 334, in get_output_data
    return_values = await _async_map_node_over_list(prompt_id, unique_id, obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, v3_data=v3_data)
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\execution.py", line 308, in _async_map_node_over_list
    await process_inputs(input_dict, i)
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\execution.py", line 296, in process_inputs
    result = f(**inputs)
             ^^^^^^^^^^^
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\comfy_api\internal\__init__.py", line 149, in wrapped_func
    return method(locked_class, **inputs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\comfy_api\latest\_io.py", line 1764, in EXECUTE_NORMALIZED
    to_return = cls.execute(*args, **kwargs)
                ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\comfy_extras\nodes_lt_audio.py", line 87, in execute
    audio = audio_vae.decode(audio_latent).to(audio_latent.device)
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\comfy\ldm\lightricks\vae\audio_vae.py", line 211, in decode
    latents = self.normalizer.denormalize(latents)
              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\comfy\ldm\lightricks\vae\audio_vae.py", line 87, in denormalize
    patched, _ = self.patchifier.patchify(latents)
                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\comfy\ldm\lightricks\symmetric_patchifier.py", line 167, in patchify
    b, _, t,_ = audio_latents.shape
    ^^^^^^^^^^
ValueError: too many values to unpack (expected 4)
Prompt executed in 116.51 seconds

---

Enabled Bypassed Nodes

got prompt
Requested to load LTXAV
Model LTXAV prepared for dynamic VRAM loading. 23838MB Staged. 1660 patches attached.
100%|████████████████████████████████████████████████████████████████████████████████████████████████| 3/3 [00:04<00:00,  1.56s/it]
Requested to load AudioVAE
loaded completely; 28630.57 MB usable, 693.46 MB loaded, full load: True
Requested to load VideoVAE
Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.
Prompt executed in 83.67 seconds
  
* * *
Trying generation on 241 length at 1920×1088
  
got prompt
Requested to load VideoVAE
Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.
Requested to load LTXAV
Model LTXAV prepared for dynamic VRAM loading. 23838MB Staged. 1660 patches attached.
100%|████████████████████████████████████████████████████████████████████████████████████████████████| 8/8 [00:41<00:00,  5.18s/it]
Requested to load VideoVAE
Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.
Requested to load LTXAV
Model LTXAV prepared for dynamic VRAM loading. 23838MB Staged. 1660 patches attached.
  0%|                                                                            | 0/3 [00:00<?, ?it/s,   Model Initializing ...  ]
33%|████████████████████▎                                        | 1/3 [01:17<02:34, 77.48s/it,  Model Initialization complete!  ]
100%|████████████████████████████████████████████████████████████████████████████████████████████████| 3/3 [01:46<00:00, 35.57s/it]
Requested to load AudioVAE
loaded completely; 28628.57 MB usable, 693.46 MB loaded, full load: True
Requested to load VideoVAE
0 models unloaded.
Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.
Windows fatal exception: access violation
Stack (most recent call first):
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\comfy\sd.py", line 902 in <lambda>
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\comfy\utils.py", line 1149 in tiled_scale_multidim
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\python_embeded\Lib\site-packages\torch\utils\_contextlib.py", line 124 in decorate_context
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\comfy\sd.py", line 903 in decode_tiled_3d
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\comfy\sd.py", line 1009 in decode_tiled
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\nodes.py", line 349 in decode
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\execution.py", line 296 in process_inputs
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\execution.py", line 308 in _async_map_node_over_list
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\execution.py", line 334 in get_output_data
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\execution.py", line 525 in execute
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\execution.py", line 758 in execute_async
  File "asyncio\events.py", line 88 in _run
  File "asyncio\base_events.py", line 1999 in _run_once
  File "asyncio\base_events.py", line 645 in run_forever
  File "asyncio\windows_events.py", line 322 in run_forever
  File "asyncio\base_events.py", line 678 in run_until_complete
  File "asyncio\runners.py", line 118 in run
  File "asyncio\runners.py", line 195 in run
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\execution.py", line 702 in execute
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\main.py", line 276 in prompt_worker
  File "threading.py", line 1012 in run
  File "threading.py", line 1075 in _bootstrap_inner
  File "threading.py", line 1032 in _bootstrap
Press any key to continue . . .

---

**Same issue

Problematic node
VAE Decode (Tiled) [Beta]
VAEDecodeTiled

Properties

TitleVAE Decode (Tiled)
ModeAlways
Color
cnr_idcomfy-core
ver0.14.1
Node name for S&RVAEDecodeTiled
enableTabsfalse
tabWidth65.000
tabXOffset10.000
hasSecondTabfalse
secondTabTextSend Back
secondTabOffset80.000
secondTabWidth65.000
ue_properties[object Object]
{"widget_ue_connectable":{},"version":"7.5.2","input_ue_unconnectable":{}}**

---

>>Trying using latest LTX2.3 I2V workflow from templates which seems updated and with newly recommended models<<
text_encoder
Gemma_3_12b_it_fp4_mixed.safetensors
LoRA (model and clip)
gemma-3-12b-it-abliterated_lora_rank64_bf16.safetensors
—Generating with default values—-

got prompt
Found quantization metadata version 1
Using MixedPrecisionOps for text encoder
CLIP/text encoder model load device: cuda:0, offload device: cpu, current: cpu, dtype: torch.float16
Found quantization metadata version 1
Detected mixed precision quantization
Using mixed precision operations
model weight dtype torch.float16, manual cast: torch.bfloat16
model_type FLUX
VAE load device: cuda:0, offload device: cpu, dtype: torch.bfloat16
no CLIP/text encoder weights in checkpoint, the text encoder model will not be loaded.
Requested to load LTXAVTEModel_
Model LTXAVTEModel_ prepared for dynamic VRAM loading. 11200MB Staged. 500 patches attached. Force pre-loaded 290 weights: 1497 KB.
Generating tokens:   0%|                                                                                   | 0/256 [00:06<?, ?it/s]
Requested to load VideoVAE
Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.
Requested to load LTXAVTEModel_
Model LTXAVTEModel_ prepared for dynamic VRAM loading. 11200MB Staged. 500 patches attached. Force pre-loaded 290 weights: 1497 KB.
Requested to load LTXAVTEModel_
Model LTXAVTEModel_ prepared for dynamic VRAM loading. 11200MB Staged. 0 patches attached. Force pre-loaded 290 weights: 1497 KB.
Requested to load LTXAV
Model LTXAV prepared for dynamic VRAM loading. 23838MB Staged. 1660 patches attached.
100%|████████████████████████████████████████████████████████████████████████████████████████████████| 8/8 [00:11<00:00,  1.38s/it]
Requested to load VideoVAE
Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.
Requested to load LTXAV
Model LTXAV prepared for dynamic VRAM loading. 23838MB Staged. 1660 patches attached.
100%|████████████████████████████████████████████████████████████████████████████████████████████████| 3/3 [00:11<00:00,  3.98s/it]
Requested to load AudioVAE
loaded completely; 28660.55 MB usable, 693.46 MB loaded, full load: True
Requested to load VideoVAE
Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.
Prompt executed in 222.04 seconds

---

—Generating at 1920×1088 241 No LoRAs —-

got prompt
Requested to load VideoVAE
Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.
Requested to load LTXAV
Model LTXAV prepared for dynamic VRAM loading. 23838MB Staged. 1660 patches attached.
100%|████████████████████████████████████████████████████████████████████████████████████████████████| 8/8 [00:37<00:00,  4.63s/it]
Requested to load VideoVAE
Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.
Requested to load LTXAV
Model LTXAV prepared for dynamic VRAM loading. 23838MB Staged. 1660 patches attached.
100%|████████████████████████████████████████████████████████████████████████████████████████████████| 3/3 [01:37<00:00, 32.54s/it]
Requested to load AudioVAE
loaded completely; 28658.62 MB usable, 693.46 MB loaded, full load: True
Requested to load VideoVAE
0 models unloaded.
Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.
Windows fatal exception: access violation
Stack (most recent call first):
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\comfy\sd.py", line 902 in <lambda>
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\comfy\utils.py", line 1149 in tiled_scale_multidim
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\python_embeded\Lib\site-packages\torch\utils\_contextlib.py", line 124 in decorate_context
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\comfy\sd.py", line 903 in decode_tiled_3d
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\comfy\sd.py", line 1009 in decode_tiled
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\nodes.py", line 349 in decode
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\execution.py", line 296 in process_inputs
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\execution.py", line 308 in _async_map_node_over_list
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\execution.py", line 334 in get_output_data
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\execution.py", line 525 in execute
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\execution.py", line 758 in execute_async
  File "asyncio\events.py", line 88 in _run
  File "asyncio\base_events.py", line 1999 in _run_once
  File "asyncio\base_events.py", line 645 in run_forever
  File "asyncio\windows_events.py", line 322 in run_forever
  File "asyncio\base_events.py", line 678 in run_until_complete
  File "asyncio\runners.py", line 118 in run
  File "asyncio\runners.py", line 195 in run
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\execution.py", line 702 in execute
  File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\main.py", line 276 in prompt_worker
  File "threading.py", line 1012 in run
  File "threading.py", line 1075 in _bootstrap_inner
  File "threading.py", line 1032 in _bootstrap
Press any key to continue . . .

---

—Issue persist reason after comfy update 0.17.0 no longer able to generate LTX2.3 at high quality 1920 at 10sec as before with an RTX 5090 and system ram 32GB—

Retrying with 1280×720 at 241

got prompt
Found quantization metadata version 1
Detected mixed precision quantization
Using mixed precision operations
model weight dtype torch.float16, manual cast: torch.bfloat16
model_type FLUX
VAE load device: cuda:0, offload device: cpu, dtype: torch.bfloat16
no CLIP/text encoder weights in checkpoint, the text encoder model will not be loaded.
Requested to load VideoVAE
Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.
Found quantization metadata version 1
Using MixedPrecisionOps for text encoder
CLIP/text encoder model load device: cuda:0, offload device: cpu, current: cpu, dtype: torch.float16
Requested to load LTXAVTEModel_
Model LTXAVTEModel_ prepared for dynamic VRAM loading. 11200MB Staged. 500 patches attached. Force pre-loaded 290 weights: 1497 KB.
Generating tokens:   0%|                                                                                   | 0/256 [00:00<?, ?it/s]
Generating tokens:   0%|                                                                                   | 0/256 [00:08<?, ?it/s]
Model LTXAVTEModel_ prepared for dynamic VRAM loading. 11200MB Staged. 500 patches attached. Force pre-loaded 290 weights: 1497 KB.
Requested to load LTXAVTEModel_
Model LTXAVTEModel_ prepared for dynamic VRAM loading. 11200MB Staged. 0 patches attached. Force pre-loaded 290 weights: 1497 KB.
Requested to load LTXAV
Model LTXAV prepared for dynamic VRAM loading. 23838MB Staged. 1660 patches attached.
  0%|                                                                            | 0/8 [00:00<?, ?it/s,   Model Initializing ...  ]
25%|████████████████████████                                                                        | 2/8 [00:05<00:16,  2.80s/it]
50%|████████████████████████████████████████████████                                                | 4/8 [00:11<00:10,  2.70s/it]
88%|████████████████████████████████████████████████████████████████████████████████████            | 7/8
100%|████████████████████████████████████████████████████████████████████████████████████████████████| 8/8 [00:21<00:00,  2.70s/it]
Requested to load VideoVAE
Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.
Requested to load LTXAV
Model LTXAV prepared for dynamic VRAM loading. 23838MB Staged. 1660 patches attached.
  0%|                                                                            | 0/3 [00:00<?, ?it/s,   Model Initializing ...  ]
100%|████████████████████████████████████████████████████████████████████████████████████████████████| 3/3 [01:10<00:00, 23.49s/it]
Requested to load AudioVAE
loaded completely; 28660.57 MB usable, 693.46 MB loaded, full load: True
Requested to load VideoVAE
0 models unloaded.
Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.
Prompt executed in 283.16 seconds

---

Memory management regression after latest update unable to generate at 1920×1088

1280×720 at 241 frames WORKS - completes successfully in 283 seconds
1920×1088 at 241 frames CRASHES - access violation in tiled VAE decode
The crash is always at sd.py line 902 in the lambda during decode_tiled_3d

[LL9Zarnek]

—Trying hunter-alpha suggested fix—

Now I can see the exact code. The decode_tiled_3d is straightforward — the decode_fn lambda is passed into tiled_scale_multidim, and the crash happens inside that lambda during the actual self.first_stage_model.decode() call on Blackwell.

Here's the fix for decode_tiled_3d:

def decode_tiled_3d(self, samples, tile_t=999, tile_x=32, tile_y=32, overlap=(1, 8, 8)):

    decode_fn = lambda a: self.first_stage_model.decode(a.to(self.vae_dtype).to(self.device)).to(dtype=self.vae_output_dtype()).contiguous()

    return self.process_output(comfy.utils.tiled_scale_multidim(samples, decode_fn, tile=(tile_t, tile_x, tile_y), overlap=overlap, upscale_amount=self.upscale_ratio, out_channels=self.output_channels, index_formulas=self.upscale_index_formula, output_device=self.output_device))

Only change: added .contiguous() at the end of the lambda chain. This forces the decoded tile into a clean memory layout before it's written back into the output tensor by tiled_scale_multidim.

---

If that still crashes

The issue may be inside the VAE's decode CUDA kernel itself (not the write-back). In that case, add an explicit device sync to flush CUDA errors before they corrupt memory:

def decode_tiled_3d(self, samples, tile_t=999, tile_x=32, tile_y=32, overlap=(1, 8, 8)):

    def safe_decode(a):

        a = a.to(self.vae_dtype).to(self.device)

        result = self.first_stage_model.decode(a)

        torch.cuda.synchronize(self.device)  # flush CUDA ops before continuing

        return result.to(dtype=self.vae_output_dtype()).contiguous()

    return self.process_output(comfy.utils.tiled_scale_multidim(samples, safe_decode, tile=(tile_t, tile_x, tile_y), overlap=overlap, upscale_amount=self.upscale_ratio, out_channels=self.output_channels, index_formulas=self.upscale_index_formula, output_device=self.output_device))

synchronize() forces all pending CUDA work to complete before the result is used, preventing the async memory corruption that causes the access violation on Blackwell.

---

Root cause

At 1920×1088, the latent is [B, C, 241, 68, 120]. With tile_t=999, each tile processes all 241 frames spatially — the 3D conv in the VideoVAE launches a massive CUDA kernel that hits a race condition / memory ordering bug in the Blackwell GPU driver for large 5D tensor operations. The 1280×720 latent [B, C, 241, 45, 80] is small enough to avoid it.

—Trying code fix1 added .contiguous()— args --disable-api-nodes --force-fp16 --fast fp16_accumulation --windows-standalone-build --disable-smart-memory --use-sage-attention --disable-mmap --enable-dynamic-vram

got prompt

Found quantization metadata version 1

Detected mixed precision quantization

Using mixed precision operations

model weight dtype torch.float16, manual cast: torch.bfloat16

model_type FLUX

VAE load device: cuda:0, offload device: cpu, dtype: torch.bfloat16

no CLIP/text encoder weights in checkpoint, the text encoder model will not be loaded.

Requested to load VideoVAE

Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.

Found quantization metadata version 1

Using MixedPrecisionOps for text encoder

CLIP/text encoder model load device: cuda:0, offload device: cpu, current: cpu, dtype: torch.float16

Requested to load LTXAVTEModel_

Model LTXAVTEModel_ prepared for dynamic VRAM loading. 11200MB Staged. 500 patches attached. Force pre-loaded 290 weights: 1497 KB.

Generating tokens:   0%|                                                                                   | 0/256 [00:00<?, ?it/s]

Generating tokens:   0%|                                                                                   | 0/256 [00:08<?, ?it/s]

Model LTXAVTEModel_ prepared for dynamic VRAM loading. 11200MB Staged. 500 patches attached. Force pre-loaded 290 weights: 1497 KB.

Requested to load LTXAVTEModel_

Model LTXAVTEModel_ prepared for dynamic VRAM loading. 11200MB Staged. 0 patches attached. Force pre-loaded 290 weights: 1497 KB.

Requested to load LTXAV

Model LTXAV prepared for dynamic VRAM loading. 23838MB Staged. 1660 patches attached.

100%|████████████████████████████████████████████████████████████████████████████████████████████████| 8/8 [00:36<00:00,  4.56s/it]

Requested to load VideoVAE

Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.

Requested to load LTXAV

Model LTXAV prepared for dynamic VRAM loading. 23838MB Staged. 1660 patches attached.

  0%|                                                                            | 0/3 [00:00<?, ?it/s,   Model Initializing ...  ]FETCH ComfyRegistry Data: 115/131

100%|████████████████████████████████████████████████████████████████████████████████████████████████| 3/3 [01:28<00:00, 29.40s/it]

Requested to load AudioVAE

loaded completely; 28660.57 MB usable, 693.46 MB loaded, full load: True

Requested to load VideoVAE

0 models unloaded.

Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.

!!! Exception during processing !!! [enforce fail at alloc_cpu.cpp:117] data. DefaultCPUAllocator: not enough memory: you tried to allocate 6041272320 bytes.

Traceback (most recent call last):

File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\execution.py", line 525, in execute

output_data, output_ui, has_subgraph, has_pending_tasks = await get_output_data(prompt_id, unique_id, obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, v3_data=v3_data)

                                                          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\execution.py", line 334, in get_output_data

return_values = await asyncmap_node_over_list(prompt_id, unique_id, obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, v3_data=v3_data)

                ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\execution.py", line 308, in asyncmap_node_over_list

await process_inputs(input_dict, i)

File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\execution.py", line 296, in process_inputs

result = f(**inputs)

         ^^^^^^^^^^^

File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\nodes.py", line 349, in decode

images = vae.decode_tiled(samples["samples"], tile_x=tile_size // compression, tile_y=tile_size // compression, overlap=overlap // compression, tile_t=temporal_size, overlap_t=temporal_overlap)

         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\comfy\sd.py", line 1009, in decode_tiled

output = self.decode_tiled_3d(samples, **args)

         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\comfy\sd.py", line 903, in decode_tiled_3d

return self.process_output(comfy.utils.tiled_scale_multidim(samples, decode_fn, tile=(tile_t, tile_x, tile_y), overlap=overlap, upscale_amount=self.upscale_ratio, out_channels=self.output_channels, index_formulas=self.upscale_index_formula, output_device=self.output_device))

                           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\python_embeded\Lib\site-packages\torch\utils_contextlib.py", line 124, in decorate_context

return func(*args, **kwargs)

       ^^^^^^^^^^^^^^^^^^^^^

File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\comfy\utils.py", line 1173, in tiled_scale_multidim

output[b:b+1] = out/out_div

                ~~~^~~~~~~~

RuntimeError: [enforce fail at alloc_cpu.cpp:117] data. DefaultCPUAllocator: not enough memory: you tried to allocate 6041272320 bytes.

Prompt executed in 285.17 seconds

—Trying code fix 2— explicit device sync to flush CUDA errors before they corrupt memory — args --disable-api-nodes --force-fp16 --fast fp16_accumulation --windows-standalone-build --disable-smart-memory --use-sage-attention --disable-mmap --enable-dynamic-vram

got prompt

Found quantization metadata version 1

Detected mixed precision quantization

Using mixed precision operations

model weight dtype torch.float16, manual cast: torch.bfloat16

model_type FLUX

VAE load device: cuda:0, offload device: cpu, dtype: torch.bfloat16

no CLIP/text encoder weights in checkpoint, the text encoder model will not be loaded.

Requested to load VideoVAE

Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.

Found quantization metadata version 1

Using MixedPrecisionOps for text encoder

CLIP/text encoder model load device: cuda:0, offload device: cpu, current: cpu, dtype: torch.float16

Requested to load LTXAVTEModel_

Model LTXAVTEModel_ prepared for dynamic VRAM loading. 11200MB Staged. 500 patches attached. Force pre-loaded 290 weights: 1497 KB.

Generating tokens:   0%|                                                                                   | 0/256 [00:00<?, ?it/s]

Generating tokens:   0%|                                                                                   | 0/256 [00:08<?, ?it/s]

Model LTXAVTEModel_ prepared for dynamic VRAM loading. 11200MB Staged. 500 patches attached. Force pre-loaded 290 weights: 1497 KB.

Requested to load LTXAVTEModel_

Model LTXAVTEModel_ prepared for dynamic VRAM loading. 11200MB Staged. 0 patches attached. Force pre-loaded 290 weights: 1497 KB.

Requested to load LTXAV

Model LTXAV prepared for dynamic VRAM loading. 23838MB Staged. 1660 patches attached.

100%|████████████████████████████████████████████████████████████████████████████████████████████████| 8/8 [00:38<00:00,  4.82s/it]

Requested to load VideoVAE

Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.

Requested to load LTXAV

Model LTXAV prepared for dynamic VRAM loading. 23838MB Staged. 1660 patches attached.

100%|████████████████████████████████████████████████████████████████████████████████████████████████| 3/3 [01:28<00:00, 29.42s/it]

Requested to load AudioVAE

loaded completely; 28660.57 MB usable, 693.46 MB loaded, full load: True

Requested to load VideoVAE

0 models unloaded.

Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.

!!! Exception during processing !!! [enforce fail at alloc_cpu.cpp:117] data. DefaultCPUAllocator: not enough memory: you tried to allocate 6041272320 bytes.

Traceback (most recent call last):

File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\execution.py", line 525, in execute

output_data, output_ui, has_subgraph, has_pending_tasks = await get_output_data(prompt_id, unique_id, obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, v3_data=v3_data)

                                                          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\execution.py", line 334, in get_output_data

return_values = await asyncmap_node_over_list(prompt_id, unique_id, obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, v3_data=v3_data)

                ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\execution.py", line 308, in asyncmap_node_over_list

await process_inputs(input_dict, i)

File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\execution.py", line 296, in process_inputs

result = f(**inputs)

         ^^^^^^^^^^^

File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\nodes.py", line 349, in decode

images = vae.decode_tiled(samples["samples"], tile_x=tile_size // compression, tile_y=tile_size // compression, overlap=overlap // compression, tile_t=temporal_size, overlap_t=temporal_overlap)

         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\comfy\sd.py", line 1019, in decode_tiled

output = self.decode_tiled_3d(samples, **args)

         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\comfy\sd.py", line 913, in decode_tiled_3d

return self.process_output(comfy.utils.tiled_scale_multidim(samples, safe_decode, tile=(tile_t, tile_x, tile_y), overlap=overlap, upscale_amount=self.upscale_ratio, out_channels=self.output_channels, index_formulas=self.upscale_index_formula, output_device=self.output_device))

                           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\python_embeded\Lib\site-packages\torch\utils_contextlib.py", line 124, in decorate_context

return func(*args, **kwargs)

       ^^^^^^^^^^^^^^^^^^^^^

File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\comfy\utils.py", line 1173, in tiled_scale_multidim

output[b:b+1] = out/out_div

                ~~~^~~~~~~~

RuntimeError: [enforce fail at alloc_cpu.cpp:117] data. DefaultCPUAllocator: not enough memory: you tried to allocate 6041272320 bytes.

Prompt executed in 288.88 seconds

Summary of attempts with code fixes stopped the memory access violation crashes but still unable to generate successfully

---

### — Trying with new Fix potentially addressing memory fragmentation preventing a contiguous allocation, PyTorch is reserving extra memory beyond what's needed—

The fix: force the output tensor onto GPU where there's VRAM available.
hunter-alpha Code:
def decode_tiled_3d(self, samples, tile_t=999, tile_x=32, tile_y=32, overlap=(1, 8, 8)):
def safe_decode(a):
a = a.to(self.vae_dtype).to(self.device)
result = self.first_stage_model.decode(a)
return result.to(dtype=self.vae_output_dtype()).contiguous()
return self.process_output(
comfy.utils.tiled_scale_multidim(
samples, safe_decode,
tile=(tile_t, tile_x, tile_y), overlap=overlap,
upscale_amount=self.upscale_ratio,
out_channels=self.output_channels,
index_formulas=self.upscale_index_formula,
output_device=self.device # ← KEY CHANGE: output on GPU, not CPU
)
)
Two changes from your original:

1. .contiguous() — prevents the Blackwell access violation

2. output_device=self.device — allocates the output tensor on GPU instead of CPU

---

Args --disable-api-nodes --force-fp16 --fast fp16_accumulation --windows-standalone-build --disable-smart-memory --use-sage-attention --disable-mmap --enable-dynamic-vram

got prompt
Found quantization metadata version 1
Detected mixed precision quantization
Using mixed precision operations
model weight dtype torch.float16, manual cast: torch.bfloat16
model_type FLUX
VAE load device: cuda:0, offload device: cpu, dtype: torch.bfloat16
no CLIP/text encoder weights in checkpoint, the text encoder model will not be loaded.
Requested to load VideoVAE
Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.
Found quantization metadata version 1
Using MixedPrecisionOps for text encoder
CLIP/text encoder model load device: cuda:0, offload device: cpu, current: cpu, dtype: torch.float16
Requested to load LTXAVTEModel_
Model LTXAVTEModel_ prepared for dynamic VRAM loading. 11200MB Staged. 500 patches attached. Force pre-loaded 290 weights: 1497 KB.
Generating tokens:   0%|                                                                                   | 0/256 [00:08<?, ?it/s]
Model LTXAVTEModel_ prepared for dynamic VRAM loading. 11200MB Staged. 500 patches attached. Force pre-loaded 290 weights: 1497 KB.
Requested to load LTXAVTEModel_
Model LTXAVTEModel_ prepared for dynamic VRAM loading. 11200MB Staged. 0 patches attached. Force pre-loaded 290 weights: 1497 KB.
Requested to load LTXAV
Model LTXAV prepared for dynamic VRAM loading. 23838MB Staged. 1660 patches attached.
100%|████████████████████████████████████████████████████████████████████████████████████████████████| 8/8 [00:36<00:00,  4.61s/it]
Requested to load VideoVAE
Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.
Requested to load LTXAV
Model LTXAV prepared for dynamic VRAM loading. 23838MB Staged. 1660 patches attached.
100%|████████████████████████████████████████████████████████████████████████████████████████████████| 3/3 [01:31<00:00, 30.34s/it]
Requested to load AudioVAE
loaded completely; 28660.57 MB usable, 693.46 MB loaded, full load: True
Requested to load VideoVAE
0 models unloaded.
Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.
Prompt executed in 291.50 seconds

---

It worked! The prompt executed successfully in 291.50 seconds without any crash. Let me confirm by reading the log:

"got prompt" - workflow started
Model loading happened successfully
Generation completed (8/8 steps, then 3/3 steps)
AudioVAE loaded
VideoVAE loaded for decode
"Prompt executed in 291.50 seconds" - SUCCESS! No crash!

What was fixed (recap)

Problem | Cause | Fix -- | -- | -- access violation (segfault) | Blackwell GPU bug with strided tensor writes | .contiguous() on decoded tiles DefaultCPUAllocator: not enough memory | 5.76 GB output tensor allocated on system RAM | output_device=self.device (GPU) Original CV_8U import errors | Broken cv2 installation | opencv-contrib-python reinstall

Keep this in sd.py line 901 - 917

  def decode_tiled_3d(self, samples, tile_t=999, tile_x=32, tile_y=32, overlap=(1, 8, 8)):
        def safe_decode(a):
               a = a.to(self.vae_dtype).to(self.device)
               result = self.first_stage_model.decode(a)
               return result.to(dtype=self.vae_output_dtype()).contiguous()
        return self.process_output(
              comfy.utils.tiled_scale_multidim(
                     samples, safe_decode,
                     tile=(tile_t, tile_x, tile_y), overlap=overlap,
                     upscale_amount=self.upscale_ratio,
                     out_channels=self.output_channels,
                     index_formulas=self.upscale_index_formula,
                     output_device=self.device
             )
      )
    

—Final checking Steps—

Retrying with old workflow with old settings and LoRAs

100%|████████████████████████████████████████████████████████████████████████████████████████████████| 3/3 [01:43<00:00, 34.50s/it]
Requested to load AudioVAE
loaded completely; 28274.57 MB usable, 693.46 MB loaded, full load: True
Requested to load VideoVAE
0 models unloaded.
Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.
Windows fatal exception: access violation
Stack (most recent call first):
File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\python_embeded\Lib\site-packages\einops_backends.py", line 96 in reshape
File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\python_embeded\Lib\site-packages\einops\einops.py", line 251 in _apply_recipe
File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\python_embeded\Lib\site-packages\einops\einops.py", line 540 in reduce
File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\python_embeded\Lib\site-packages\einops\einops.py", line 616 in rearrange
File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\comfy\ldm\lightricks\vae\causal_video_autoencoder.py", line 1030 in unpatchify
File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\comfy\ldm\lightricks\vae\causal_video_autoencoder.py", line 566 in forward_orig
File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\comfy\ldm\lightricks\vae\causal_video_autoencoder.py", line 572 in forward
File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1787 in _call_impl
File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1776 in _wrapped_call_impl
File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\comfy\ldm\lightricks\vae\causal_video_autoencoder.py", line 1205 in decode
File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\comfy\sd.py", line 904 in safe_decode
File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\comfy\utils.py", line 1149 in tiled_scale_multidim
File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\python_embeded\Lib\site-packages\torch\utils_contextlib.py", line 124 in decorate_context
File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\comfy\sd.py", line 908 in decode_tiled_3d
File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\comfy\sd.py", line 1022 in decode_tiled
File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\nodes.py", line 349 in decode
File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\execution.py", line 296 in process_inputs
File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\execution.py", line 308 in _async_map_node_over_list
File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\execution.py", line 334 in get_output_data
File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\execution.py", line 525 in execute
File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\execution.py", line 758 in execute_async
File "asyncio\events.py", line 88 in _run
File "asyncio\base_events.py", line 1999 in _run_once
File "asyncio\base_events.py", line 645 in run_forever
File "asyncio\windows_events.py", line 322 in run_forever
File "asyncio\base_events.py", line 678 in run_until_complete
File "asyncio\runners.py", line 118 in run
File "asyncio\runners.py", line 195 in run
File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\execution.py", line 702 in execute
File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\main.py", line 276 in prompt_worker
File "threading.py", line 1012 in run
File "threading.py", line 1075 in _bootstrap_inner
File "threading.py", line 1032 in _bootstrap

—Retry and Sticking to new workflow and Trying to migrate our 5 LoRAs and settings to the new workflow—

100%|████████████████████████████████████████████████████████████████████████████████████████████████| 8/8 [00:42<00:00, 5.28s/it]
Requested to load VideoVAE
Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.
Requested to load LTXAV
Model LTXAV prepared for dynamic VRAM loading. 23838MB Staged. 1660 patches attached.
100%|████████████████████████████████████████████████████████████████████████████████████████████████| 3/3 [01:31<00:00, 30.56s/it]
Requested to load AudioVAE
loaded completely; 28660.62 MB usable, 693.46 MB loaded, full load: True
Requested to load VideoVAE
0 models unloaded.
Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.
Windows fatal exception: access violation
Stack (most recent call first):
File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\comfy\sd.py", line 905 in safe_decode
File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\comfy\utils.py", line 1149 in tiled_scale_multidim
File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\python_embeded\Lib\site-packages\torch\utils_contextlib.py", line 124 in decorate_context
File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\comfy\sd.py", line 908 in decode_tiled_3d
File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\comfy\sd.py", line 1022 in decode_tiled
File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\nodes.py", line 349 in decode
File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\execution.py", line 296 in process_inputs
File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\execution.py", line 308 in _async_map_node_over_list
File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\execution.py", line 334 in get_output_data
File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\execution.py", line 525 in execute
File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\execution.py", line 758 in execute_async
File "asyncio\events.py", line 88 in _run
File "asyncio\base_events.py", line 1999 in _run_once
File "asyncio\base_events.py", line 645 in run_forever
File "asyncio\windows_events.py", line 322 in run_forever
File "asyncio\base_events.py", line 678 in run_until_complete
File "asyncio\runners.py", line 118 in run
File "asyncio\runners.py", line 195 in run
File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\execution.py", line 702 in execute
File "I:\Ai\ComfyUI_RTX5090\ComfyUI-Easy-Install\ComfyUI\main.py", line 276 in prompt_worker
File "threading.py", line 1012 in run
File "threading.py", line 1075 in _bootstrap_inner
File "threading.py", line 1032 in _bootstrap
Press any key to continue . . .

Strange could it be too many LoRAs? but it worked before so very unlikely…wait but the 5 LoRA is new…lets try loading only 4 LoRAs then

got prompt
Found quantization metadata version 1
Detected mixed precision quantization
Using mixed precision operations
model weight dtype torch.float16, manual cast: torch.bfloat16
model_type FLUX
VAE load device: cuda:0, offload device: cpu, dtype: torch.bfloat16
no CLIP/text encoder weights in checkpoint, the text encoder model will not be loaded.
Requested to load VideoVAE
Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.
FETCH ComfyRegistry Data: 30/131
Found quantization metadata version 1
Using MixedPrecisionOps for text encoder
CLIP/text encoder model load device: cuda:0, offload device: cpu, current: cpu, dtype: torch.float16
Requested to load LTXAVTEModel_
Model LTXAVTEModel_ prepared for dynamic VRAM loading. 11200MB Staged. 500 patches attached. Force pre-loaded 290 weights: 1497 KB.
Generating tokens: 0%| | 0/256 [00:00<?, ?it/s]FETCH ComfyRegistry Data: 35/131
FETCH ComfyRegistry Data: 40/131
Generating tokens: 0%| | 0/256 [00:08<?, ?it/s]
Model LTXAVTEModel_ prepared for dynamic VRAM loading. 11200MB Staged. 500 patches attached. Force pre-loaded 290 weights: 1497 KB.
Requested to load LTXAVTEModel_
Model LTXAVTEModel_ prepared for dynamic VRAM loading. 11200MB Staged. 0 patches attached. Force pre-loaded 290 weights: 1497 KB.
FETCH ComfyRegistry Data: 45/131
Requested to load LTXAV
Model LTXAV prepared for dynamic VRAM loading. 23838MB Staged. 1660 patches attached.
FETCH ComfyRegistry Data: 50/131
0%| | 0/8 [00:00<?, ?it/s, Model Initializing ... ]FETCH ComfyRegistry Data: 55/131
FETCH ComfyRegistry Data: 60/131
FETCH ComfyRegistry Data: 65/131
FETCH ComfyRegistry Data: 70/131
FETCH ComfyRegistry Data: 75/131
FETCH ComfyRegistry Data: 80/131
FETCH ComfyRegistry Data: 85/131
FETCH ComfyRegistry Data: 90/131
12%|███████▋ | 1/8 [00:55<06:25, 55.05s/it, Model Initialization complete! ]FETCH ComfyRegistry Data: 95/131
25%|████████████████████████ | 2/8 [00:09<00:27, 4.66s/it]FETCH ComfyRegistry Data: 100/131
50%|████████████████████████████████████████████████ | 4/8 [00:19<00:18, 4.70s/it]FETCH ComfyRegistry Data: 105/131
62%|████████████████████████████████████████████████████████████ | 5/8 [00:24<00:14, 4.68s/it]FETCH ComfyRegistry Data: 110/131
88%|████████████████████████████████████████████████████████████████████████████████████ | 7/8 [00:33<00:04, 4.69s/it]FETCH ComfyRegistry Data: 115/131
100%|████████████████████████████████████████████████████████████████████████████████████████████████| 8/8 [00:38<00:00, 4.78s/it]
FETCH ComfyRegistry Data: 120/131
FETCH ComfyRegistry Data: 125/131
Requested to load VideoVAE
Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.
Requested to load LTXAV
Model LTXAV prepared for dynamic VRAM loading. 23838MB Staged. 1660 patches attached.
FETCH ComfyRegistry Data: 130/131
0%| | 0/3 [00:00<?, ?it/s, Model Initializing ... ]FETCH ComfyRegistry Data [DONE]
[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]
[ComfyUI-Manager] All startup tasks have been completed.
100%|████████████████████████████████████████████████████████████████████████████████████████████████| 3/3 [01:36<00:00, 32.18s/it]
Requested to load AudioVAE
loaded completely; 28660.57 MB usable, 693.46 MB loaded, full load: True
Requested to load VideoVAE
0 models unloaded.
Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.
Prompt executed in 322.71 seconds
Exception in callback _ProactorBasePipeTransport._call_connection_lost(None)
handle: <Handle _ProactorBasePipeTransport._call_connection_lost(None)>
Traceback (most recent call last):
File "asyncio\events.py", line 88, in _run
File "asyncio\proactor_events.py", line 165, in _call_connection_lost
ConnectionResetError: [WinError 10054] An existing connection was forcibly closed by the remote host

—
This confirms it 4 LoRA is max to avoid hitting strange memory issues
take into consideration the 5th Lora size is 1GB while the rest between 4 LoRAs 600-500MB

VRAM Budget Breakdown

Component | VRAM -- | -- LTXAV main model | 23,838 MB Text encoder (Gemma) | 11,200 MB VideoVAE | 1,384 MB AudioVAE | 693 MB 4 LoRAs (500-600 MB each) | ~2,200 MB 5th LoRA (1 GB) | +1,000 MB Total with 5 LoRAs | ~40,315 MB ❌ Total with 4 LoRAs | ~39,315 MB ✅ (barely)

With dynamic VRAM management, the text encoder is offloaded during diffusion, so actual peak is closer to ~29 GB — but adding that 5th LoRA pushes it past the 32 GB limit.