Memory management changes and leaks after v0.30.2->v0.33.1 update

[LordLocksley]

Expected Behavior

Upgraded Comfyui installation from 0.30.2 to 0.33.1, then to 0.33.2

Expected behavior is that same workflows with the same models and settings with the same nodes will behave the same memory wise, and generate in a comparable amount of time.

This is not happening, see Actual Behavior section below.

Actual Behavior

After updating Comfyui from 0.30.2 to 0.33.1 (and issue persists after upgrading to latest 0.33.2), some memory management changes must have happened in one of the intermediate versions, and workflows that worked perfectly fine on 0.30.2 now fill all available RAM and VRAM and need SSD swap and run extremely slow (not due to inherent slower inference but due to disk thrashing) or even run OOM with Comfyui v0.33.1/2. No other updates to the system or Portable Comfyui installation/node packs besides the base Comfyui ones.
Update done using ComfyUI Manager.

Happens mostly on LTX 2.3 derivative 10Eros, LTX 2.3 base and to a lesser extent on Minimax H3 (albeit this may or may not be related). Also tested LTX2.5 after updating to 0.33.1, and got similar results.

Besides of being more memory hungry right from the start, it also seems to be a memory leak somewhere, as running the same workflow 4-6 times (same exact run, just queueing it several times for reseeding), gives OOM after 3/4 tries or so, when I could queue 8-12 videos with the same exact settings on 0.30.2 without OOM crashing (probably more, but I don't think I ever queued more than 12 at once).
Albeit originally noticed with LTX 2.3, this also seem to affect LTX 2.5, albeit I didn't test LTX 2.5 on the previous 0.30.2, only after upgrading to get support for it. First few runs fills up all memory available, degrading in performance as SSD swap happens, then subsequent runs OOMs even on conservative settings.

Workflows used are those:
For LTX 2.3/10Eros: https://huggingface.co/LokkenJP/10EROS_1.5_fp8_exp_learned/tree/main/workflows
This model: https://huggingface.co/LokkenJP/10EROS_1.5_int8_convrot_exp_learned (tested other quants or flavors of LTX 2.3, all int8 convrot, same result)
For LTX 2.5: The ComfyUI standard workflow. Standard distilled int8 convrot model straight from Lightricks repo.
For H3: The ComfyUI standard workflow. Int8 convrot quant from https://huggingface.co/Abiray/Minimax-H3-nvfp4-INT4-INT8-Convrot/tree/main

For LTX using 0.7-0.8mpx videos of 20s or so duration. Never spilled RAM over limit to SSD swap before on 0.30.2, now fill ALL memory just on the first run and starts swapping to the disk drive soon after, then OOMs on subsequent runs.
For H3 using 0.6Mpx and 12-14s. Same behavior. on 0.30.2 same settings allowed this setup repeatedly without OOMs.

Desktop machine with AMD 5800X3D 32Gb Ram, 4080 RTX 16 Gb VRAM, Windows 11, Portable Comfyui installation.
Python version: 3.13.12 (tags/v3.13.12:1cbe481, Feb 3 2026, 18:22:25) [MSC v.1944 64 bit (AMD64)]
pytorch version: 2.12.0+cu130

Using Sage Attention 2.2.0+cu130torch2.10.0andhigher.post6 but I was also using it (same exact version) on 0.30.2 without any memory issue.

Steps to Reproduce

Just queue several LTX 2.3 videos with a int8 convrot model in a row with settings just within the RAM constraint of 0.30.2. when doing the same on 0.33.2 instead, monitor memory load. Quite easy to spot the difference.

In the log attached I queued the same workflow a few times one after another, same settings just different random generated seed, notice how the first run took 451.09 seconds, second one 307.58 seconds as models were already loaded, then third Prompt executed in 359.41 seconds, 4th one spiked to 00:10:05 due to heavy SSD swapping, and fifth run crashed OOM.

Debug Logs

f:\ComfyUI\ComfyUI_windows_portable_new>.\python_embeded\python.exe -s ComfyUI\main.py --windows-standalone-build --use-sage-attention --fast
[INFO] setup plugin alembic.autogenerate.schemas
[INFO] setup plugin alembic.autogenerate.tables
[INFO] setup plugin alembic.autogenerate.types
[INFO] setup plugin alembic.autogenerate.constraints
[INFO] setup plugin alembic.autogenerate.defaults
[INFO] setup plugin alembic.autogenerate.comments
[INFO] Adding extra search path checkpoints e:\ComfyUI\model_storage\models\checkpoints
[INFO] Adding extra search path loras e:\ComfyUI\model_storage\models\loras
[INFO] Adding extra search path vae e:\ComfyUI\model_storage\models\vae
[INFO] Adding extra search path text_encoders e:\ComfyUI\model_storage\models\text_encoders
[INFO] Adding extra search path text_encoders e:\ComfyUI\model_storage\models\clip
[INFO] Adding extra search path diffusion_models e:\ComfyUI\model_storage\models\unet
[INFO] Adding extra search path diffusion_models e:\ComfyUI\model_storage\models\diffusion_models
[INFO] Adding extra search path embeddings e:\ComfyUI\model_storage\models\embeddings
[INFO] Adding extra search path diffusers e:\ComfyUI\model_storage\models\diffusers
[INFO] Adding extra search path upscale_models e:\ComfyUI\model_storage\models\upscale_models
[INFO] Adding extra search path latent_upscale_models e:\ComfyUI\model_storage\models\latent_upscale_models
[START] Security scan
[DONE] Security scan
## ComfyUI-Manager: installing dependencies done.
** ComfyUI startup time: 2026-08-20 20:26:19.353
** Platform: Windows
** Python version: 3.13.12 (tags/v3.13.12:1cbe481, Feb  3 2026, 18:22:25) [MSC v.1944 64 bit (AMD64)]
** Python executable: f:\ComfyUI\ComfyUI_windows_portable_new\python_embeded\python.exe
** ComfyUI Path: f:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI
** ComfyUI Base Folder Path: f:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI
** User directory: F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\user
** ComfyUI-Manager config path: F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\user\__manager\config.ini
** Log path: F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\user\comfyui.log
[INFO]
Prestartup times for custom nodes:
[INFO]    0.0 seconds: F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\custom_nodes\rgthree-comfy
[INFO]    0.0 seconds: F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\custom_nodes\comfyui-easy-use
[INFO]    4.2 seconds: F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\custom_nodes\ComfyUI-Manager
[INFO]
[INFO] Found comfy_kitchen backend hip: {'available': False, 'disabled': False, 'unavailable_reason': 'PyTorch ROCm/HIP runtime not available', 'capabilities': []}
[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 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] 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] Checkpoint files will always be loaded safely.
[INFO] Total VRAM 16376 MB, total RAM 32694 MB
[INFO] pytorch version: 2.12.0+cu130
[INFO] Enabled fp16 accumulation.
[INFO] Set vram state to: NORMAL_VRAM
[INFO] Device: cuda:0 NVIDIA GeForce RTX 4080 : native
[INFO] Using async weight offloading with 2 streams
[INFO] Enabled pinned memory 13077.0
[INFO] Using sage attention
[INFO] aimdo: src-win/cuda-detour.c:38:INFO:aimdo_setup_hooks: installing 6 hooks
[INFO] aimdo: src/control.c:262:INFO:comfy-aimdo NVML pressure enabled
[INFO] aimdo: src-win/shmem-detect.c:80:INFO:comfy-aimdo WDDM adapter match: NVIDIA GeForce RTX 4080 runtime_luid=00000000:0000eba2 dxgi_luid=00000000:0000eba2
[INFO] aimdo: src/control.c:277:INFO:comfy-aimdo inited for GPU: NVIDIA GeForce RTX 4080 (VRAM: 16375 MB)
[INFO] DynamicVRAM support detected and enabled
[INFO] Python version: 3.13.12 (tags/v3.13.12:1cbe481, Feb  3 2026, 18:22:25) [MSC v.1944 64 bit (AMD64)]
[INFO] ComfyUI version: 0.33.2
[INFO] comfy-aimdo version: 0.4.13
[INFO] comfy-kitchen version: 0.2.31
[INFO] comfyui-frontend-package version: 1.49.6
[INFO] comfyui-workflow-templates version: 0.11.43
[INFO] comfyui-embedded-docs version: 0.5.9
[INFO] comfy-kitchen version: 0.2.31
[INFO] comfy-aimdo version: 0.4.13
[INFO] [Prompt Server] web root: f:\ComfyUI\ComfyUI_windows_portable_new\python_embeded\Lib\site-packages\comfyui_frontend_package\static
[INFO] Asset seeder disabled
[INFO] No OpenGL_accelerate module loaded: No module named 'OpenGL_accelerate'
[AcademiaSD Loader] Loading comfyui_AcademiaSD v1.9.7
[AcademiaSD Loader] Adding ComfyUI root path to sys.path: F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI
[AcademiaSD Loader] Successfully pre-imported the 'comfy' module.
[F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\custom_nodes\comfy-mtb] | INFO -> loaded 102 nodes successfuly
[F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\custom_nodes\comfy-mtb] | INFO -> Some nodes (6) could not be loaded. This can be ignored, but go to http://127.0.0.1:8188/mtb if you want more information.
Web extensions folder found at F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\web\extensions\ComfyLiterals
[ComfyUI-Easy-Use] server: v1.3.6 Loaded
[ComfyUI-Easy-Use] web root: F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\custom_nodes\comfyui-easy-use\web_version/v2 Loaded
[INFO] ComfyUI-GGUF: Allowing full torch compile
[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-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 Revision: 473 [7cee3ceb] *DETACHED | Released on '2026-08-17'
[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
[INFO] ComfyUI-GGUF: Allowing full torch compile
------------------------------------------
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
------------------------------------------
[WARNING] Unable to parse pyproject.toml due to lack dependency pydantic-settings, please run 'pip install -r requirements.txt': Invalid statement (at line 1, column 1)

Initializing ControlAltAI Nodes
[INFO] Using sage attention
(RES4LYF) Init
(RES4LYF) Importing beta samplers.
(RES4LYF) Importing legacy samplers.

[rgthree-comfy] Loaded 48 exciting 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 ✅
💡 Optional: pip install flash-attn
🔧 Conv3d workaround active: PyTorch 2.12.0, cuDNN 92000 (fixing VAE 3x memory bug)
📊 Initial CUDA memory: 14.68GB free / 15.99GB total
[LazyPrompt] Loaded 11 model skill(s) from Model_Skills/
[INFO]
Import times for custom nodes:
[INFO]    0.0 seconds: F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\custom_nodes\websocket_image_save.py
[INFO]    0.0 seconds: F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\custom_nodes\comfyui_nvidia_rtx_nodes
[INFO]    0.0 seconds: F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\custom_nodes\ComfyUI-HiresFix-Ultra-AllInOne
[INFO]    0.0 seconds: F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\custom_nodes\Nvidia_RTX_Nodes_ComfyUI
[INFO]    0.0 seconds: F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\custom_nodes\comfyui-gps-supplements
[INFO]    0.0 seconds: F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\custom_nodes\ComfyLiterals
[INFO]    0.0 seconds: F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\custom_nodes\ComfyUI-GGUF
[INFO]    0.0 seconds: F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\custom_nodes\ComfyUI-VFI
[INFO]    0.0 seconds: F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\custom_nodes\comfyui-mxtoolkit
[INFO]    0.0 seconds: F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\custom_nodes\comfyui-various
[INFO]    0.0 seconds: F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\custom_nodes\ComfyUI-Conditioning-Rebalance
[INFO]    0.0 seconds: F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\custom_nodes\ComfyMath
[INFO]    0.0 seconds: F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\custom_nodes\comfyui-custom-scripts
[INFO]    0.0 seconds: F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\custom_nodes\ComfyUI-Zlycoris
[INFO]    0.0 seconds: F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\custom_nodes\controlaltai-nodes
[INFO]    0.0 seconds: F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\custom_nodes\10S_Nodes
[INFO]    0.0 seconds: F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\custom_nodes\gguf
[INFO]    0.0 seconds: F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\custom_nodes\ComfyUI-VideoHelperSuite
[INFO]    0.1 seconds: F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\custom_nodes\rgthree-comfy
[INFO]    0.1 seconds: F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\custom_nodes\comfyui-kjnodes
[INFO]    0.1 seconds: F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\custom_nodes\ComfyUI-LTXVideo
[INFO]    0.1 seconds: F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\custom_nodes\Vsaan212-workflow-utilities
[INFO]    0.2 seconds: F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\custom_nodes\comfy-mtb
[INFO]    0.2 seconds: F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\custom_nodes\comfyui-minimax-h3-audio-T8-main
[INFO]    0.2 seconds: F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\custom_nodes\comfyui_layerstyle
[INFO]    0.3 seconds: F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\custom_nodes\ComfyUI-Manager
[INFO]    0.3 seconds: F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\custom_nodes\ComfyUI-Impact-Pack
[INFO]    0.4 seconds: F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\custom_nodes\seedvr2_videoupscaler
[INFO]    0.4 seconds: F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\custom_nodes\RES4LYF
[INFO]    0.5 seconds: F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\custom_nodes\ComfyUI_Comfyroll_CustomNodes
[INFO]    0.6 seconds: F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\custom_nodes\comfyui-rtx-simple
[INFO]    0.9 seconds: F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\custom_nodes\ComfyUI_StarNodes
[INFO]    0.9 seconds: F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\custom_nodes\academiasd
[INFO]    1.2 seconds: F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\custom_nodes\ComfyUI-S3-IO
[INFO]    3.2 seconds: F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\custom_nodes\comfyui-easy-use
[INFO]
[INFO] Context impl SQLiteImpl.
[INFO] Will assume non-transactional DDL.
[INFO] Using RAM pressure cache.
[INFO] Starting server

[INFO] To see the GUI go to: http://127.0.0.1:8188
[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: /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\ComfyUI_windows_portable_new\ComfyUI\custom_nodes\comfy-mtb] | INFO -> Found multiple match, we will pick the last F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\models\upscale_models
['e:\\ComfyUI\\model_storage\\models\\upscale_models', 'F:\\ComfyUI\\ComfyUI_windows_portable_new\\ComfyUI\\models\\upscale_models']
[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.
[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.
[INFO] got prompt
[INFO] VAE load device: cuda:0, offload device: cpu, dtype: torch.float32
[INFO] Found quantization metadata version 1
[INFO] Detected mixed precision quantization
[INFO] Using mixed precision operations
[INFO] Native ops: float8_e4m3fn, float8_e5m2, convrot_w4a4, asym_w4a8_int8, int8_tensorwise , emulated ops: nvfp4, mxfp8
[INFO] model weight dtype torch.bfloat16, manual cast: torch.bfloat16
[INFO] model_type FLUX
[INFO] VAE load device: cuda:0, offload device: cpu, dtype: torch.bfloat16
[WARNING] no CLIP/text encoder weights in checkpoint, the text encoder model will not be loaded.
[INFO] Requested to load VideoVAE
[INFO] Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.
[INFO] Found quantization metadata version 1
[INFO] Using MixedPrecisionOps for text encoder
[WARNING] Missing weight for layer vision_model.encoder.layers.0.self_attn.q_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.0.self_attn.k_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.0.self_attn.v_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.0.self_attn.out_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.0.mlp.fc1
[WARNING] Missing weight for layer vision_model.encoder.layers.0.mlp.fc2
[WARNING] Missing weight for layer vision_model.encoder.layers.1.self_attn.q_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.1.self_attn.k_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.1.self_attn.v_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.1.self_attn.out_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.1.mlp.fc1
[WARNING] Missing weight for layer vision_model.encoder.layers.1.mlp.fc2
[WARNING] Missing weight for layer vision_model.encoder.layers.2.self_attn.q_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.2.self_attn.k_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.2.self_attn.v_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.2.self_attn.out_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.2.mlp.fc1
[WARNING] Missing weight for layer vision_model.encoder.layers.2.mlp.fc2
[WARNING] Missing weight for layer vision_model.encoder.layers.3.self_attn.q_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.3.self_attn.k_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.3.self_attn.v_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.3.self_attn.out_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.3.mlp.fc1
[WARNING] Missing weight for layer vision_model.encoder.layers.3.mlp.fc2
[WARNING] Missing weight for layer vision_model.encoder.layers.4.self_attn.q_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.4.self_attn.k_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.4.self_attn.v_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.4.self_attn.out_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.4.mlp.fc1
[WARNING] Missing weight for layer vision_model.encoder.layers.4.mlp.fc2
[WARNING] Missing weight for layer vision_model.encoder.layers.5.self_attn.q_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.5.self_attn.k_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.5.self_attn.v_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.5.self_attn.out_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.5.mlp.fc1
[WARNING] Missing weight for layer vision_model.encoder.layers.5.mlp.fc2
[WARNING] Missing weight for layer vision_model.encoder.layers.6.self_attn.q_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.6.self_attn.k_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.6.self_attn.v_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.6.self_attn.out_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.6.mlp.fc1
[WARNING] Missing weight for layer vision_model.encoder.layers.6.mlp.fc2
[WARNING] Missing weight for layer vision_model.encoder.layers.7.self_attn.q_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.7.self_attn.k_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.7.self_attn.v_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.7.self_attn.out_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.7.mlp.fc1
[WARNING] Missing weight for layer vision_model.encoder.layers.7.mlp.fc2
[WARNING] Missing weight for layer vision_model.encoder.layers.8.self_attn.q_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.8.self_attn.k_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.8.self_attn.v_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.8.self_attn.out_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.8.mlp.fc1
[WARNING] Missing weight for layer vision_model.encoder.layers.8.mlp.fc2
[WARNING] Missing weight for layer vision_model.encoder.layers.9.self_attn.q_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.9.self_attn.k_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.9.self_attn.v_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.9.self_attn.out_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.9.mlp.fc1
[WARNING] Missing weight for layer vision_model.encoder.layers.9.mlp.fc2
[WARNING] Missing weight for layer vision_model.encoder.layers.10.self_attn.q_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.10.self_attn.k_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.10.self_attn.v_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.10.self_attn.out_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.10.mlp.fc1
[WARNING] Missing weight for layer vision_model.encoder.layers.10.mlp.fc2
[WARNING] Missing weight for layer vision_model.encoder.layers.11.self_attn.q_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.11.self_attn.k_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.11.self_attn.v_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.11.self_attn.out_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.11.mlp.fc1
[WARNING] Missing weight for layer vision_model.encoder.layers.11.mlp.fc2
[WARNING] Missing weight for layer vision_model.encoder.layers.12.self_attn.q_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.12.self_attn.k_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.12.self_attn.v_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.12.self_attn.out_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.12.mlp.fc1
[WARNING] Missing weight for layer vision_model.encoder.layers.12.mlp.fc2
[WARNING] Missing weight for layer vision_model.encoder.layers.13.self_attn.q_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.13.self_attn.k_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.13.self_attn.v_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.13.self_attn.out_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.13.mlp.fc1
[WARNING] Missing weight for layer vision_model.encoder.layers.13.mlp.fc2
[WARNING] Missing weight for layer vision_model.encoder.layers.14.self_attn.q_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.14.self_attn.k_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.14.self_attn.v_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.14.self_attn.out_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.14.mlp.fc1
[WARNING] Missing weight for layer vision_model.encoder.layers.14.mlp.fc2
[WARNING] Missing weight for layer vision_model.encoder.layers.15.self_attn.q_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.15.self_attn.k_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.15.self_attn.v_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.15.self_attn.out_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.15.mlp.fc1
[WARNING] Missing weight for layer vision_model.encoder.layers.15.mlp.fc2
[WARNING] Missing weight for layer vision_model.encoder.layers.16.self_attn.q_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.16.self_attn.k_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.16.self_attn.v_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.16.self_attn.out_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.16.mlp.fc1
[WARNING] Missing weight for layer vision_model.encoder.layers.16.mlp.fc2
[WARNING] Missing weight for layer vision_model.encoder.layers.17.self_attn.q_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.17.self_attn.k_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.17.self_attn.v_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.17.self_attn.out_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.17.mlp.fc1
[WARNING] Missing weight for layer vision_model.encoder.layers.17.mlp.fc2
[WARNING] Missing weight for layer vision_model.encoder.layers.18.self_attn.q_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.18.self_attn.k_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.18.self_attn.v_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.18.self_attn.out_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.18.mlp.fc1
[WARNING] Missing weight for layer vision_model.encoder.layers.18.mlp.fc2
[WARNING] Missing weight for layer vision_model.encoder.layers.19.self_attn.q_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.19.self_attn.k_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.19.self_attn.v_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.19.self_attn.out_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.19.mlp.fc1
[WARNING] Missing weight for layer vision_model.encoder.layers.19.mlp.fc2
[WARNING] Missing weight for layer vision_model.encoder.layers.20.self_attn.q_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.20.self_attn.k_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.20.self_attn.v_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.20.self_attn.out_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.20.mlp.fc1
[WARNING] Missing weight for layer vision_model.encoder.layers.20.mlp.fc2
[WARNING] Missing weight for layer vision_model.encoder.layers.21.self_attn.q_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.21.self_attn.k_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.21.self_attn.v_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.21.self_attn.out_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.21.mlp.fc1
[WARNING] Missing weight for layer vision_model.encoder.layers.21.mlp.fc2
[WARNING] Missing weight for layer vision_model.encoder.layers.22.self_attn.q_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.22.self_attn.k_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.22.self_attn.v_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.22.self_attn.out_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.22.mlp.fc1
[WARNING] Missing weight for layer vision_model.encoder.layers.22.mlp.fc2
[WARNING] Missing weight for layer vision_model.encoder.layers.23.self_attn.q_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.23.self_attn.k_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.23.self_attn.v_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.23.self_attn.out_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.23.mlp.fc1
[WARNING] Missing weight for layer vision_model.encoder.layers.23.mlp.fc2
[WARNING] Missing weight for layer vision_model.encoder.layers.24.self_attn.q_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.24.self_attn.k_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.24.self_attn.v_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.24.self_attn.out_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.24.mlp.fc1
[WARNING] Missing weight for layer vision_model.encoder.layers.24.mlp.fc2
[WARNING] Missing weight for layer vision_model.encoder.layers.25.self_attn.q_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.25.self_attn.k_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.25.self_attn.v_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.25.self_attn.out_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.25.mlp.fc1
[WARNING] Missing weight for layer vision_model.encoder.layers.25.mlp.fc2
[WARNING] Missing weight for layer vision_model.encoder.layers.26.self_attn.q_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.26.self_attn.k_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.26.self_attn.v_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.26.self_attn.out_proj
[WARNING] Missing weight for layer vision_model.encoder.layers.26.mlp.fc1
[WARNING] Missing weight for layer vision_model.encoder.layers.26.mlp.fc2
[WARNING] 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.embeddings.position_embedding.weight', 'vision_model.encoder.layers.0.layer_norm1.weight', 'vision_model.encoder.layers.0.layer_norm1.bias', 'vision_model.encoder.layers.0.layer_norm2.weight', 'vision_model.encoder.layers.0.layer_norm2.bias', 'vision_model.encoder.layers.1.layer_norm1.weight', 'vision_model.encoder.layers.1.layer_norm1.bias', 'vision_model.encoder.layers.1.layer_norm2.weight', 'vision_model.encoder.layers.1.layer_norm2.bias', 'vision_model.encoder.layers.2.layer_norm1.weight', 'vision_model.encoder.layers.2.layer_norm1.bias', 'vision_model.encoder.layers.2.layer_norm2.weight', 'vision_model.encoder.layers.2.layer_norm2.bias', 'vision_model.encoder.layers.3.layer_norm1.weight', 'vision_model.encoder.layers.3.layer_norm1.bias', 'vision_model.encoder.layers.3.layer_norm2.weight', 'vision_model.encoder.layers.3.layer_norm2.bias', 'vision_model.encoder.layers.4.layer_norm1.weight', 'vision_model.encoder.layers.4.layer_norm1.bias', 'vision_model.encoder.layers.4.layer_norm2.weight', 'vision_model.encoder.layers.4.layer_norm2.bias', 'vision_model.encoder.layers.5.layer_norm1.weight', 'vision_model.encoder.layers.5.layer_norm1.bias', 'vision_model.encoder.layers.5.layer_norm2.weight', 'vision_model.encoder.layers.5.layer_norm2.bias', 'vision_model.encoder.layers.6.layer_norm1.weight', 'vision_model.encoder.layers.6.layer_norm1.bias', 'vision_model.encoder.layers.6.layer_norm2.weight', 'vision_model.encoder.layers.6.layer_norm2.bias', 'vision_model.encoder.layers.7.layer_norm1.weight', 'vision_model.encoder.layers.7.layer_norm1.bias', 'vision_model.encoder.layers.7.layer_norm2.weight', 'vision_model.encoder.layers.7.layer_norm2.bias', 'vision_model.encoder.layers.8.layer_norm1.weight', 'vision_model.encoder.layers.8.layer_norm1.bias', 'vision_model.encoder.layers.8.layer_norm2.weight', 'vision_model.encoder.layers.8.layer_norm2.bias', 'vision_model.encoder.layers.9.layer_norm1.weight', 'vision_model.encoder.layers.9.layer_norm1.bias', 'vision_model.encoder.layers.9.layer_norm2.weight', 'vision_model.encoder.layers.9.layer_norm2.bias', 'vision_model.encoder.layers.10.layer_norm1.weight', 'vision_model.encoder.layers.10.layer_norm1.bias', 'vision_model.encoder.layers.10.layer_norm2.weight', 'vision_model.encoder.layers.10.layer_norm2.bias', 'vision_model.encoder.layers.11.layer_norm1.weight', 'vision_model.encoder.layers.11.layer_norm1.bias', 'vision_model.encoder.layers.11.layer_norm2.weight', 'vision_model.encoder.layers.11.layer_norm2.bias', 'vision_model.encoder.layers.12.layer_norm1.weight', 'vision_model.encoder.layers.12.layer_norm1.bias', 'vision_model.encoder.layers.12.layer_norm2.weight', 'vision_model.encoder.layers.12.layer_norm2.bias', 'vision_model.encoder.layers.13.layer_norm1.weight', 'vision_model.encoder.layers.13.layer_norm1.bias', 'vision_model.encoder.layers.13.layer_norm2.weight', 'vision_model.encoder.layers.13.layer_norm2.bias', 'vision_model.encoder.layers.14.layer_norm1.weight', 'vision_model.encoder.layers.14.layer_norm1.bias', 'vision_model.encoder.layers.14.layer_norm2.weight', 'vision_model.encoder.layers.14.layer_norm2.bias', 'vision_model.encoder.layers.15.layer_norm1.weight', 'vision_model.encoder.layers.15.layer_norm1.bias', 'vision_model.encoder.layers.15.layer_norm2.weight', 'vision_model.encoder.layers.15.layer_norm2.bias', 'vision_model.encoder.layers.16.layer_norm1.weight', 'vision_model.encoder.layers.16.layer_norm1.bias', 'vision_model.encoder.layers.16.layer_norm2.weight', 'vision_model.encoder.layers.16.layer_norm2.bias', 'vision_model.encoder.layers.17.layer_norm1.weight', 'vision_model.encoder.layers.17.layer_norm1.bias', 'vision_model.encoder.layers.17.layer_norm2.weight', 'vision_model.encoder.layers.17.layer_norm2.bias', 'vision_model.encoder.layers.18.layer_norm1.weight', 'vision_model.encoder.layers.18.layer_norm1.bias', 'vision_model.encoder.layers.18.layer_norm2.weight', 'vision_model.encoder.layers.18.layer_norm2.bias', 'vision_model.encoder.layers.19.layer_norm1.weight', 'vision_model.encoder.layers.19.layer_norm1.bias', 'vision_model.encoder.layers.19.layer_norm2.weight', 'vision_model.encoder.layers.19.layer_norm2.bias', 'vision_model.encoder.layers.20.layer_norm1.weight', 'vision_model.encoder.layers.20.layer_norm1.bias', 'vision_model.encoder.layers.20.layer_norm2.weight', 'vision_model.encoder.layers.20.layer_norm2.bias', 'vision_model.encoder.layers.21.layer_norm1.weight', 'vision_model.encoder.layers.21.layer_norm1.bias', 'vision_model.encoder.layers.21.layer_norm2.weight', 'vision_model.encoder.layers.21.layer_norm2.bias', 'vision_model.encoder.layers.22.layer_norm1.weight', 'vision_model.encoder.layers.22.layer_norm1.bias', 'vision_model.encoder.layers.22.layer_norm2.weight', 'vision_model.encoder.layers.22.layer_norm2.bias', 'vision_model.encoder.layers.23.layer_norm1.weight', 'vision_model.encoder.layers.23.layer_norm1.bias', 'vision_model.encoder.layers.23.layer_norm2.weight', 'vision_model.encoder.layers.23.layer_norm2.bias', 'vision_model.encoder.layers.24.layer_norm1.weight', 'vision_model.encoder.layers.24.layer_norm1.bias', 'vision_model.encoder.layers.24.layer_norm2.weight', 'vision_model.encoder.layers.24.layer_norm2.bias', 'vision_model.encoder.layers.25.layer_norm1.weight', 'vision_model.encoder.layers.25.layer_norm1.bias', 'vision_model.encoder.layers.25.layer_norm2.weight', 'vision_model.encoder.layers.25.layer_norm2.bias', 'vision_model.encoder.layers.26.layer_norm1.weight', 'vision_model.encoder.layers.26.layer_norm1.bias', 'vision_model.encoder.layers.26.layer_norm2.weight', 'vision_model.encoder.layers.26.layer_norm2.bias', 'vision_model.post_layernorm.weight', 'vision_model.post_layernorm.bias']
[INFO] CLIP/text encoder model load device: cuda:0, offload device: cpu, current: cpu, dtype: torch.float16
[INFO] Requested to load LTXAVTEModel_
[INFO] Model LTXAVTEModel_ prepared for dynamic VRAM loading. 14396MB Staged. 0 patches attached. Force pre-loaded 562 weights: 2280 KB.
[INFO] Model LTXAVTEModel_ prepared for dynamic VRAM loading. 14396MB Staged. 0 patches attached. Force pre-loaded 562 weights: 2280 KB.
[LTX Ref] v2 patched _prepare_positional_embeddings on LTXAVModel (preserving original from LTXAVModel)
[LTX Ref] Global patches applied to LTXAVModel.
[INFO] Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.
[INFO] Requested to load LTXAV
[INFO] 0 models unloaded.
[INFO] Model LTXAV prepared for dynamic VRAM loading. 24897MB Staged. 1660 patches attached. Force pre-loaded 608 weights: 3303 KB.
  0%|                                                               | 0/11 [00:00<?, ?it/s,   Model Initializing ...  ]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.
100%|██████████████████████████████████████████████████████████████████████████████████| 11/11 [01:02<00:00,  5.65s/it]
[INFO] Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.
[INFO] Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.
[INFO] Requested to load LTXAV
[INFO] Model LTXAV prepared for dynamic VRAM loading. 24897MB Staged. 1660 patches attached. Force pre-loaded 608 weights: 3303 KB.
100%|████████████████████████████████████████████████████████████████████████████████████| 3/3 [01:45<00:00, 35.03s/it]
[INFO] Requested to load AudioVAE
[INFO] loaded completely;  693.46 MB loaded, full load: True
[INFO] 0 models unloaded.
[INFO] Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.
[INFO] Prompt executed in 451.09 seconds
[INFO] got prompt
[INFO] Requested to load LTXAV
[INFO] Model LTXAV prepared for dynamic VRAM loading. 24897MB Staged. 1660 patches attached. Force pre-loaded 608 weights: 3303 KB.
  0%|                                                               | 0/11 [00:00<?, ?it/s,   Model Initializing ...  ][INFO] got prompt
100%|██████████████████████████████████████████████████████████████████████████████████| 11/11 [00:50<00:00,  4.63s/it]
[INFO] Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.
[INFO] Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.
[INFO] Requested to load LTXAV
[INFO] Model LTXAV prepared for dynamic VRAM loading. 24897MB Staged. 1660 patches attached. Force pre-loaded 608 weights: 3303 KB.
100%|████████████████████████████████████████████████████████████████████████████████████| 3/3 [01:47<00:00, 35.68s/it]
[INFO] Requested to load AudioVAE
[INFO] loaded completely;  693.46 MB loaded, full load: True
[INFO] 0 models unloaded.
[INFO] Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.
[INFO] Prompt executed in 307.58 seconds
[INFO] Requested to load LTXAV
[INFO] Model LTXAV prepared for dynamic VRAM loading. 24897MB Staged. 1660 patches attached. Force pre-loaded 608 weights: 3303 KB.
100%|██████████████████████████████████████████████████████████████████████████████████| 11/11 [00:49<00:00,  4.50s/it]
[INFO] Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.
[INFO] Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.
[INFO] Requested to load LTXAV
[INFO] Model LTXAV prepared for dynamic VRAM loading. 24897MB Staged. 1660 patches attached. Force pre-loaded 608 weights: 3303 KB.
100%|████████████████████████████████████████████████████████████████████████████████████| 3/3 [01:48<00:00, 36.02s/it]
[INFO] Requested to load AudioVAE
[INFO] loaded completely;  693.46 MB loaded, full load: True
[INFO] 0 models unloaded.
[INFO] Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.
[INFO] Prompt executed in 359.41 seconds
[ERROR] Exception in callback _ProactorBasePipeTransport._call_connection_lost()
handle: <Handle _ProactorBasePipeTransport._call_connection_lost()>
Traceback (most recent call last):
  File "asyncio\events.py", line 89, in _run
  File "asyncio\proactor_events.py", line 165, in _call_connection_lost
ConnectionResetError: [WinError 10054] Se ha forzado la interrupción de una conexión existente por el host remoto
[INFO] got prompt
[INFO] Requested to load LTXAV
[INFO] Model LTXAV prepared for dynamic VRAM loading. 24897MB Staged. 1660 patches attached. Force pre-loaded 608 weights: 3303 KB.
100%|██████████████████████████████████████████████████████████████████████████████████| 11/11 [01:25<00:00,  7.73s/it]
[INFO] Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.
[INFO] Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.
[INFO] Requested to load LTXAV
[INFO] Model LTXAV prepared for dynamic VRAM loading. 24897MB Staged. 1660 patches attached. Force pre-loaded 608 weights: 3303 KB.
100%|████████████████████████████████████████████████████████████████████████████████████| 3/3 [01:56<00:00, 38.70s/it]
[INFO] Requested to load AudioVAE
[INFO] loaded completely;  693.46 MB loaded, full load: True
[INFO] 0 models unloaded.
[INFO] Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.
[INFO] Prompt executed in 00:10:05
[ERROR] Exception in callback _ProactorBasePipeTransport._call_connection_lost()
handle: <Handle _ProactorBasePipeTransport._call_connection_lost()>
Traceback (most recent call last):
  File "asyncio\events.py", line 89, in _run
  File "asyncio\proactor_events.py", line 165, in _call_connection_lost
ConnectionResetError: [WinError 10054] Se ha forzado la interrupción de una conexión existente por el host remoto
[INFO] got prompt
[INFO] Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.
[INFO] Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.
[INFO] Requested to load LTXAV
[INFO] Model LTXAV prepared for dynamic VRAM loading. 24897MB Staged. 1660 patches attached. Force pre-loaded 608 weights: 3303 KB.
100%|██████████████████████████████████████████████████████████████████████████████████| 11/11 [01:55<00:00, 10.53s/it]
[INFO] Requested to load VideoVAE
[INFO] Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.
[INFO] Model VideoVAE prepared for dynamic VRAM loading. 1384MB Staged. 0 patches attached.
[INFO] Requested to load LTXAV
[INFO] 0 models unloaded.
[INFO] Model LTXAV prepared for dynamic VRAM loading. 24897MB Staged. 1660 patches attached. Force pre-loaded 608 weights: 3303 KB.
  0%|                                                                | 0/3 [00:00<?, ?it/s,   Model Initializing ...  ][INFO] aimdo: src/vrambuf.c:91:INFO:VRAM Allocation failed (OOM)
  0%|                                                                | 0/3 [02:09<?, ?it/s,   Model Initializing ...  ]
[ERROR] !!! Exception during processing !!! VRAM grow failed: 644274176 bytes
[ERROR] Traceback (most recent call last):
  File "f:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\execution.py", line 545, 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 "f:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\execution.py", line 344, 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 "f:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\execution.py", line 318, in _async_map_node_over_list
    await process_inputs(input_dict, i)
  File "f:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\execution.py", line 306, in process_inputs
    result = f(**inputs)
  File "f:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\comfy_api\internal\__init__.py", line 149, in wrapped_func
    return method(locked_class, **inputs)
  File "f:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\comfy_api\latest\_io.py", line 1990, in EXECUTE_NORMALIZED
    to_return = cls.execute(*args, **kwargs)
  File "F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\comfy_extras\nodes_custom_sampler.py", line 777, in execute
    samples = comfy.sample.sample_custom(model, noise, cfg, sampler, sigmas, positive, negative, latent_image, noise_mask=noise_mask, callback=callback, disable_pbar=disable_pbar, seed=noise_seed)
  File "f:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\comfy\sample.py", line 84, in sample_custom
    samples = comfy.samplers.sample(model, noise, positive, negative, cfg, model.load_device, sampler, sigmas, model_options=model.model_options, latent_image=latent_image, denoise_mask=noise_mask, callback=callback, disable_pbar=disable_pbar, seed=seed)
  File "f:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\comfy\samplers.py", line 1353, in sample
    return cfg_guider.sample(noise, latent_image, sampler, sigmas, denoise_mask, callback, disable_pbar, seed)
           ~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "f:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\comfy\samplers.py", line 1335, in sample
    output = executor.execute(noise, latent_image, sampler, sigmas, denoise_mask, callback, disable_pbar, seed, latent_shapes=latent_shapes)
  File "f:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\comfy\patcher_extension.py", line 113, in execute
    return self.original(*args, **kwargs)
           ~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
  File "f:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\comfy\samplers.py", line 1262, in outer_sample
    output = self.inner_sample(noise, latent_image, device, sampler, sigmas, denoise_mask, callback, disable_pbar, seed, latent_shapes=latent_shapes)
  File "f:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\comfy\samplers.py", line 1237, in inner_sample
    samples = executor.execute(self, sigmas, extra_args, callback, noise, latent_image, denoise_mask, disable_pbar)
  File "f:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\comfy\patcher_extension.py", line 113, in execute
    return self.original(*args, **kwargs)
           ~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
  File "f:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\comfy\samplers.py", line 1005, in sample
    samples = self.sampler_function(model_k, noise, sigmas, extra_args=extra_args, callback=k_callback, disable=disable_pbar, **self.extra_options)
  File "f:\ComfyUI\ComfyUI_windows_portable_new\python_embeded\Lib\site-packages\torch\utils\_contextlib.py", line 124, in decorate_context
    return func(*args, **kwargs)
  File "F:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\comfy\k_diffusion\sampling.py", line 1288, in sample_euler_ancestral_cfg_pp
    denoised = model(x, sigmas[i] * s_in, **extra_args)
  File "f:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\comfy\samplers.py", line 640, in __call__
    out = self.inner_model(x, sigma, model_options=model_options, seed=seed)
  File "f:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\comfy\samplers.py", line 1208, in __call__
    return self.outer_predict_noise(*args, **kwargs)
           ~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
  File "f:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\comfy\samplers.py", line 1215, in outer_predict_noise
    ).execute(x, timestep, model_options, seed)
      ~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "f:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\comfy\patcher_extension.py", line 113, in execute
    return self.original(*args, **kwargs)
           ~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
  File "f:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\comfy\samplers.py", line 1218, in predict_noise
    return sampling_function(self.inner_model, x, timestep, self.conds.get("negative", None), self.conds.get("positive", None), self.cfg, model_options=model_options, seed=seed)
  File "f:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\comfy\samplers.py", line 620, in sampling_function
    out = calc_cond_batch(model, conds, x, timestep, model_options)
  File "f:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\comfy\samplers.py", line 211, in calc_cond_batch
    return _calc_cond_batch_outer(model, conds, x_in, timestep, model_options)
  File "f:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\comfy\samplers.py", line 219, in _calc_cond_batch_outer
    return executor.execute(model, conds, x_in, timestep, model_options)
           ~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "f:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\comfy\patcher_extension.py", line 113, in execute
    return self.original(*args, **kwargs)
           ~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
  File "f:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\comfy\samplers.py", line 335, in _calc_cond_batch
    output = model.apply_model(input_x, timestep_, **c).chunk(batch_chunks)
             ~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^
  File "f:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\comfy\model_base.py", line 204, in apply_model
    return comfy.patcher_extension.WrapperExecutor.new_class_executor(
           ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
    ...<2 lines>...
        comfy.patcher_extension.get_all_wrappers(comfy.patcher_extension.WrappersMP.APPLY_MODEL, transformer_options)
        ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
    ).execute(x, t, c_concat, c_crossattn, control, transformer_options, **kwargs)
    ~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "f:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\comfy\patcher_extension.py", line 113, in execute
    return self.original(*args, **kwargs)
           ~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
  File "f:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\comfy\model_base.py", line 248, in _apply_model
    model_output = self.diffusion_model(xc, t, context=context, control=control, transformer_options=transformer_options, **extra_conds)
  File "f:\ComfyUI\ComfyUI_windows_portable_new\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1778, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
           ~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
  File "f:\ComfyUI\ComfyUI_windows_portable_new\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1789, in _call_impl
    return forward_call(*args, **kwargs)
  File "f:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\comfy\ldm\lightricks\av_model.py", line 1097, in forward
    return super().forward(
           ~~~~~~~~~~~~~~~^
        x,
        ^^
    ...<6 lines>...
        **kwargs,
        ^^^^^^^^^
    )
    ^
  File "f:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\comfy\ldm\lightricks\model.py", line 966, in forward
    return comfy.patcher_extension.WrapperExecutor.new_class_executor(
           ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
    ...<4 lines>...
        ),
        ~~
    ).execute(x, timestep, context, attention_mask, frame_rate, transformer_options, keyframe_idxs, denoise_mask=denoise_mask, **kwargs)
    ~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "f:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\comfy\patcher_extension.py", line 113, in execute
    return self.original(*args, **kwargs)
           ~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
  File "f:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\comfy\ldm\lightricks\model.py", line 1019, in _forward
    x = self._process_transformer_blocks(
        x, context, attention_mask, timestep, pe,
    ...<2 lines>...
        **merged_args,
    )
  File "f:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\comfy\ldm\lightricks\av_model.py", line 942, in _process_transformer_blocks
    comfy.model_prefetch.prefetch_queue_pop(prefetch_queue, vx.device, block)
    ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "f:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\comfy\model_prefetch.py", line 104, in prefetch_queue_pop
    offload_stream, fully_faulted = comfy.ops.cast_modules_with_vbar(comfy_modules, None, device, None, True, return_faulted=True)
                                    ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "f:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\comfy\ops.py", line 208, in cast_modules_with_vbar
    xfer_dest = get_cast_buffer(dest_size)
  File "f:\ComfyUI\ComfyUI_windows_portable_new\ComfyUI\comfy\ops.py", line 161, in get_cast_buffer
    buffer = comfy_aimdo.torch.aimdo_to_tensor(cast_buffer.get(buffer_size, cast_buffer_offset), device)
                                               ~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "f:\ComfyUI\ComfyUI_windows_portable_new\python_embeded\Lib\site-packages\comfy_aimdo\vram_buffer.py", line 49, in get
    raise RuntimeError(f"VRAM grow failed: {required_size} bytes")
RuntimeError: VRAM grow failed: 644274176 bytes

[INFO] Prompt executed in 573.89 seconds

Other

No response

[Code2Collapse]

+1 in this. using ubuntu 24.04 and its freezing like crazy

[LordLocksley]

Tried to review the changes between 0.30.2 to 0.33.1, both main code and dependency changes, focusing on LTX paths.
Before completing further diagnostic tests, the current list of potentially relevant changes is:

1. LTX inference optimizations: 15989f87 introduced fused rms_adaln and linear_input_act execution throughout LTX. These change kernel ordering and intermediate-tensor lifetimes. This could be related because the regression is most consistently observed with LTX, and another user already reported reduced LTX memory headroom after this change. The change looks sound, but worth exploring nevertheless.

2. AIMDO 0.4.11→0.4.13: ComfyUI updated the dependency in ba1d0efc. AIMDO 0.4.13 includes GPU copies on the default CUDA stream, which could interact with concurrent offload-stream transfers and heavily patched INT8 ConvRot models.

3. prefetch/casting bookkeeping: efd4e951 refactored model_prefetch.py for CUDA-graph support and changed some generic queue, fault and cleanup bookkeeping. Normal LTX execution does not appear to enable the new CUDA-graph branch, but the shared prefetch path was still modified. The related casting lifetime refactor is cbbc9dab.

4. pinned host-buffer reuse: There is a related report of AIMDO host buffers growing until their reservation is exhausted. That reproduction is not identical and I found no specific v0.30.2→v0.33.2 commit introducing the underlying buffer design, but it is relevant given the observed progressive system-RAM and swap growth.

5. comfy-kitchen 0.2.26→0.2.31: The reference was updated but the relevant CUDA implementations of rms_adaln, int8_linear, and INT8 ConvRot dequantization are unchanged in the package comparison. Most additions concern W4A8, neighborhood attention and optional comfy-kitchen INT8 attention; these runs use external SageAttention so should not be ralated. Worth mentioning though.

This list is intended only as a starting reference of things that have changed in between versions. I'm not claiming any of those are the actual cause.

According to the helper tool I used to review this, I will try now to make some runs with --disable-async-offload flag and later with --disable-dynamic-vram flag. Some users claim those flags fixed their issues, and at the very least that could help in pinpointing a bit more the issue. Also will try to undo locally the 15989f87 commit, and see if this helps with the issue

[monstari]

Not sure if its coming from same cause but I'll share my experience too.

Currently when I refresh the page or press R to reload node definitions, the RAM usage goes crazy. From 96 GB to slowly filling my entire RAM, and while those happen I can't run workflows... It just freezes. It only happens after I use comfy a while later tho.

[Code2Collapse]

i also started to notice that the entire UI is going bonkers including shortcuts not working at all?