✨One-click Pod available on:✨
🟣 Runpod ComfyUI 0.19.0 CUDA 12.8 for 5090
🟣 Runpod ComfyUI 0.19.0 CUDA 12.4 for 4090
🟡 VastAI ComfyUI 0.19.0 CUDA 13.0 for 5090
🟡 VastAI ComfyUI 0.19.0 CUDA 13.0 for 4090
Just click on the link, choose a Video Card and the Template will install all you need + all my workflows.
Wan 2.2 Models are not included, you can install it using Civicomfy or ComfyUI-HuggingFace directly inside ComfyUI
OneClick-I2V-Story on Runpod Basic Tutorial
https://limewire.com/d/2TAty#No5eoN7WhU
☕️ buymeacoffee
IMPORTANT:
If you install RES4LYF node it will broke the MoEKSampler, to use it you have to use the KSampler included in that node.
🔥 02/28/26 UPDATE 🔥
✨ NEW NODES ✨
EASY MODEL DOWNLOAD FROM HUGGINGFACE
cd ComfyUI/custom_nodes
git clone https://github.com/huchukato/ComfyUI-HuggingFace.gitRIFE INTERPOLATION WITH TENSORRT
with Auto Install (CUDA 12/CUDA 13) and Auto Model Download
cd ComfyUI/custom_nodes
git clone https://github.com/huchukato/ComfyUI-RIFE-TensorRT-Auto.gitUPSCALER WITH TENSORRT
with Auto Install (CUDA 12/CUDA 13) and Auto Model Download
cd ComfyUI/custom_nodes
git clone https://github.com/huchukato/ComfyUI-Upscaler-TensorRT-Auto.git✨MY QWEN3-VL NODE IS FINALLY IN THE MANAGER✨
Just search for "QwenVL-Mod" to install it
or dowload it from GitHub: QwenVL-Mod: Enhanced Vision-Language
ComfyUI-QwenVL-Mod - Enhanced Vision-Language with WAN 2.2
Version 2.2.4 (2026/03/13) - 🎬 Critical I2V Timeline Fixes & NSFW Presets Optimization
🌟 What is ComfyUI-QwenVL-Mod?
A powerful enhanced vision-language node for ComfyUI that combines Qwen3-VL models with professional WAN 2.2 video generation workflows. Features multilingual support, visual style detection, and NSFW capabilities for professional AI content creation.
Think: "Your all-in-one solution for intelligent prompt enhancement and video generation with cutting-edge AI models!"
🎬 Key Features
🚀 WAN 2.2 Video Generation
Text-to-Video (T2V): Professional 5-second video generation
Image-to-Video (I2V): Advanced image animation with style detection
Story Generation: 20-second continuous videos with 4 narrative segments
Storyboard Workflows: Seamless storyboard-to-storyboard generation
Cinematic Video: Professional cinematography specifications
🌐 Enhanced Capabilities
Multilingual Support: Process prompts from any language (Italian, English, etc.)
Visual Style Detection: 12+ artistic styles (anime, 3D, pixel art, puppet animation, etc.)
Smart Prompt Caching: Performance optimization with Fixed Seed Mode
GGUF Backend: Efficient local model inference with quantization support
NSFW Support: Comprehensive content generation without restrictions
🧠 Intelligent Features
Auto-Prompt Enhancement: Automatically enhance user prompts for optimal generation
Professional Cinematography: Built-in specifications for lighting, camera angles, shot types
Timeline Structure: Precise 5-second timeline with frame-by-frame descriptions
Keep Last Prompt: Generate once, preserve results while changing inputs
🎯 What's New in v2.2.4 - CRITICAL I2V TIMELINE FIXES
🚨 Major I2V Timeline (20s) Fixes
✅ Style Coherence: Fixed AI changing anime→realism mid-sequence
✅ Character Stability: Fixed characters disappearing/appearing incorrectly
✅ Natural Lighting: Fixed AI adding artificial lights not in image
✅ Timeline Structure: Fixed continuous numbering (6,7,8...) instead of 0-5 restart
✅ Format Consistency: Fixed missing parentheses and unwanted labels
✅ Output Format: Each prompt starts directly with timeline markers
🔧 NSFW Presets Optimization
✅ Complete Specifications: All 8 NSFW presets now include full NSFW descriptions
✅ Emoji Display: Restored proper emoji rendering (🍿🎥🎬📖)
✅ Clear Instructions: Removed confusing recommendations from presets
✅ User Guide: Token settings guide created for workflow optimization
📋 Technical Improvements
✅ Timeline Markers: Correct
(At X seconds: ...)format for all 4 prompts✅ Character Continuity: Natural progression without forced artificial presence
✅ Lighting Rules: Logical progression instead of absolute prohibitions
✅ Style Detection: Consistent style application across all timeline segments
🎯 Model Recommendations
Qwen3-VL-8B: Recommended for I2V Timeline (20s) complex sequences
Qwen3-VL-4B: Sufficient for I2V Scene (5s) single prompts
Token Settings: 2048+ for 20s timeline, 1024+ for 5s prompts
🎯 What's New in v2.2.3
CUDA 13 Compatibility: Fixed crashes caused by conflicting unload operations
Parameter Cleanup: Removed redundant unload_after_run from all nodes
Bug Fixes: Resolved "missing required positional argument" errors
Memory Management: Streamlined VRAM cleanup with VRAM Cleanup node
Documentation: Updated all README files with new memory features
Credits: Added community credits for feedback and testing
🎯 What's New in v2.2.2
🚀 Critical T2V/I2V Workflow Fixes
Batch Processing: Fixed critical T2V → GGUF issue with batch images
Frame Detection: Added automatic batch detection and individual frame processing
Video Support: Enhanced video frame processing with proper shape handling
Debug Enhanced: Comprehensive logging for batch processing troubleshooting
🔄 Same Model Reuse Fix
Conflict Resolution: Fixed crash when using same model between T2V and I2V nodes
Memory Management: Enhanced cleanup with CUDA synchronization and timing
Signature Mismatch: Resolved different signature patterns between nodes
Aggressive Cleanup: Forced complete VRAM cleanup before model reload
🔧 keep_model_loaded Enhancement
Missing Parameter: Added keep_model_loaded to PromptEnhancer node
Consistent Behavior: Both GGUF and PromptEnhancer now have identical memory management
Conditional Cleanup: Proper cleanup based on keep_model_loaded setting
User Control: Full control over memory usage vs performance
🚨 CRITICAL BUG FIXES - v2.2.4
🎬 I2V Timeline (20s) - COMPLETELY FIXED
Before v2.2.4:
❌ Anime style changed to realism mid-sequence
❌ Characters disappeared/appeared randomly
❌ AI added artificial lights not in image
❌ Timeline numbering: 6,7,8... instead of 0-5 restart
❌ Missing parentheses and unwanted "Prompt 1:" labels
After v2.2.4:
✅ Perfect Style Coherence: Anime stays anime, realism stays realism
✅ Character Stability: Same characters throughout all 4 prompts
✅ Natural Lighting: Only lights visible in image, logical progression
✅ Correct Timeline: Each prompt uses 0-5 seconds format
✅ Clean Output: Proper
(At X seconds: ...)format, no labels
🔥 NSFW Presets - ENHANCED & FIXED
✅ Complete Specifications: All 8 presets with full explicit descriptions
✅ Emoji Display: Proper 🍿🎥🎬📖 icons (no more unicode codes)
✅ User-Friendly: Removed confusing technical recommendations
✅ Token Guide: Workflow note for optimal settings
🎯 Result: Perfect I2V Timeline generation every time!
🎬 WAN 2.2 Story Workflow - Revolutionary AI Storytelling
📖 AI Story Generation
4-Segment Videos: Automatic 20-second videos (4 × 5-second segments)
Narrative Continuity: Perfect story flow between segments
NSFW Support: Enhanced adult content generation
Timeline-Free: Natural storytelling without time markers
🔄 Smart Auto-Split
Story Split Node: Intelligent prompt separation technology
Auto-Detection: Handles any separator format automatically
4-Output Guarantee: Always produces exactly 4 prompts
Debug Mode: Built-in troubleshooting information
📦 Installation
Requirements
ComfyUI: v0.13.0+
GPU: 8GB+ VRAM (16GB+ recommended)
System: Windows/Linux/Mac
Python: 3.10+ (or use provided Docker environment)
Docker/Cloud Ready
RunPod: Pre-configured templates available
VastAI: Optimized instances ready
Local: Docker support included
Quick Install
Download: ComfyUI-QwenVL-Mod (latest version)
Extract to
ComfyUI/custom_nodes/ComfyUI-QwenVL-ModRestart ComfyUI
Load included workflows
🎮 Usage Examples
Basic Image-to-Video
Load WAN2.2-I2V-AutoPrompt.json
Upload your image
Select model (HF or GGUF)
Generate enhanced video
Basic Text-to-Video
Load WAN2.2-T2V-AutoPrompt.json
Input your text prompt
Select model (HF or GGUF)
Generate enhanced video
Image-to-Video with Style
Load WAN2.2-I2V-AutoPrompt.json
Upload your image
Enable style detection
Generate animated video
AI Story Generation
Load WAN2.2-I2V-AutoPrompt-Story.json
Input your story idea
Auto-split into 4 segments
Generate 20-second story video
🔧 Technical Specifications
⚡ Performance
Context: 65,536 tokens (8B models)
Memory: Optimized VRAM usage
Stability: Crash-free operation
Speed: Fast generation times
🎨 Model Support
Qwen3-VL 4B: 7 GGUF variants (2.38GB-4.28GB)
Qwen3-VL 8B: 7 GGUF variants (4.8GB-8.71GB)
HF Models: Josiefed and official variants
Quantization: Q4_K_S, Q5_K_S for VRAM efficiency
🌐 Multilingual Capabilities
Input Languages: Any language supported
Auto-Translation: Automatic translation to optimized English
Style Detection: Works with multilingual prompts
Cultural Adaptation: Context-aware prompt enhancement
🎯 Included Workflows
🍿 WAN 2.2 Presets
🍿 Wan 2.2 I2V Timeline: Image-to-video with timeline structure
🎥 Wan 2.2 I2V Scene: Image-to-video with professional specs
🍿 Wan 2.2 T2V Timeline: Text-to-video with timeline structure
🎥 Wan 2.2 T2V Scene: Text-to-video with professional specs
🎬 Wan 2.2 I2V Timeline: 20-second multi-segment continuity
📖 Wan 2.2 I2V Scene: 20-second single scene with cinematography
🔥 Advanced Features
NSFW Enhancement: Uncensored content generation
Professional Lighting: 8 light types + 9 qualities
Camera Control: 6 shot types + 5 compositions
Color Grading: 4 tone options
🎨 Visual Style Detection
Automatically detects and enhances:
Photorealistic style - Realistic lighting, natural textures, lifelike details
Anime style - Japanese animation aesthetics with vibrant colors
Cartoon style - Bold outlines, flat colors, exaggerated expressions
3D animation style - Computer-generated 3D rendering characteristics
Fantasy style - Magical elements, ethereal lighting, imaginative atmosphere
Artistic portrait style - Painterly qualities, artistic brushwork, stylized composition
🔥 NSFW Content Support
Enhanced Generation
Explicit Content: Uncensored adult descriptions
Detailed Scenes: 8-12 sentences per segment
Natural Progression: Smooth story flow
Style Adaptation: Automatic visual style matching
Quality: Consistent characters & scenes
Professional Applications
Adult Content: Industry-standard generation
Artistic Nudity: Classical art styles
Educational: Anatomy and artistic study
Creative: Artistic expression
🚀 Why Choose ComfyUI-QwenVL-Mod?
🎬 For Content Creators
Storytelling: Create compelling narratives
Efficiency: One prompt → complete video
Quality: Professional video output
Flexibility: Any genre, any style
🔥 For NSFW Content
Explicit: Uncensored generation
Detailed: Rich scene descriptions
Continuous: Smooth story flow
Natural: Realistic progression
⚡ For Power Users
Customizable: Easy to modify
Extendable: Add more segments
Integrable: Works with existing setups
Optimized: Maximum performance
🌟 What Makes This Special?
First: Complete AI story system with vision enhancement
Smart: Intelligent prompt splitting and enhancement
Complete: End-to-end solution from text to video
Optimized: Performance-tuned for professional use
Ready: Works out-of-the-box with included workflows
🎬 Create Amazing AI Videos Today!
Transform your ideas into stunning videos with the power of Qwen3-VL vision enhancement and WAN 2.2 video generation.
Perfect for creators, artists, and professionals looking for the ultimate AI video enhancement tool! 🌟
Built with ❤️ for the ComfyUI community
🔶 All the images used to create the videos in the gallery are generated with PimpMyPony 🔶
The new version is out today, with a brand new anime style ✨
WORKFLOWS TESTED ON:
ComfyUI 0.17.1
Python 3.12.12
Pytorch 2.9.1 + CUDA 13.0
🟣 Credits
This workflows are intended to be used with the models by taek75799 as they follow the structure of Dynamic Prompts you can find under these models:
WAN 2.2 Enhanced NSFW | SVI | camera prompt adherence (Lightning Edition) I2V and T2V fp8 GGUF
🟣 Other tested models:
Smooth Mix Wan 2.2 14B (I2V/T2V) by DigitalPastel | GGUF versions by Santodan
Wan2.2-Remix (T2V&I2V) by FX_FeiHou | GGUF versions by Santodan
Thanks to all the users who are commenting and helping me improve the workflows ❤️
🔖 T2V-I2V AUTOPROMPT
🛑 Experimental WF 🛑
Start with a T2V prompt and extend the generated video with I2V
This workflow requires both T2V and I2V Wan 2.2 Models
🔖 SVI I2V AUTOPROMPT
Thanks to taek75799 for his models ❤️
SVI LORAS:
LIGHTX2V LoRaS are not included in the model
🔖 FULL I2V AUTOPROMPT
Complete workflow that includes:
Long Video Generation [from 5 to 20 seconds]
Auto Prompting [Qwen3-VL]
Upscale [2xLexicaRRDBNet and Tensorrt]
Frame Interpolation [30fps and 60fps for img2vid | 24fps and 50fps for MMAudio]
MMAudio [NSFW Unlocked]
📝 AUTO PROMPT
Prompt Description Box [Multilanguage]: Just write your idea and the LLM will do the rest, formatting the prompt in the dynamic prompt format used in the Wan 2.2 models
Final Prompt Preview: Shows up the final prompt
🔶 QWEN3-VL NODE FOR GGUF MODELS 🔶
🔖 To use the Qwen3-VL GGUF Quantized models you have to install llama-cpp-python
If you are not comfortable in manual install the llama, just go with the normal version inside Full-I2V-LongVideo
🛑 STOP COMFYUI
📂 Activate the ComfyUI Virtual Enviroment
In your ComfyUI root installation folder type:
on Windows:
Command Prompt:
\venv\Scripts\activate.bator Power Shell:
\venv\Scripts\Activate.ps1on Linux:
. /venv/bin/activateIf you use ComfyUI Desktop:
Click on Console and then on Terminal
⬇️ Install llama-cpp-python
pip install --upgrade --force-reinstall --no-cache-dir "llama-cpp-python @ git+https://github.com/JamePeng/llama-cpp-python.git"🔄 Restart ComfyUI and enjoy
📝 SWITCH CLIP
By default the WF uses the GGUF node to load quantized Clip, if you wanna switch to the NSFW Clip model, you have to bypass the GGUF node and connect the other Clip loader to the "Set_CLIP"
ACCELERATION:
Triton is disabled by default, you can enable it by opening the first Subgraph
Inside the workflow you will find all the links to download the models you will need
That's all, hope you enjoy ^^
Description
🛑 Experimental WF 🛑
Start with a T2V prompt and extend the generated video with I2V
This workflow requires both T2V and I2V Wan 2.2 Models
FAQ
Comments (52)
how to update your nodes to get a new preset "Wan Extended Storyboard: Timeline + continuity + professional spec" ? I have your node, but before the 6.02 update
Open the ComfyUI Manager and click on "Update All", will update all the custom nodes, including mine
Just running your updated Non GUFF model and the following error is shown -- Failed to validate prompt for output 1327:
* WanMoeKSamplerAdvanced 1252:1284:
- Return type mismatch between linked nodes: scheduler, received_type(['simple', 'sgm_uniform', 'karras', 'exponential', 'ddim_uniform', 'beta', 'normal', 'linear_quadratic', 'kl_optimal', 'bong_tangent']) mismatch input_type(['simple', 'sgm_uniform', 'karras', 'exponential', 'ddim_uniform', 'beta', 'normal', 'linear_quadratic', 'kl_optimal', 'bong_tangent', 'beta57'])
Which scheduler you see in the dropdown in the Selectors node?
Euler for the sampler selector and simple for the schedular selector
@pyeeater283 If you open the dropdown of the scheduler selector what scheduler you have in the list? simple, sgm uniform, karras, beta etc
@huchukato simple, sgm_uniform, karras, exponential, ddim_uniform, betas, normal, linear_quadratic, kl_optimal, bong_tangent, beta57
@pyeeater283 ok you have a list of scheduler that the KSampler do not support, I don't know why, to avoid the error disable the scheduler selector or delete it, go inside the Subgraphs and manual set the scheduler (use simple and euler as a sampler, they are good for almost all the models) in the WanMoESampler
OK , found the issue , something to do with RES4LYF inserting itself in the list , deleted that custom node and its resolved. Although the model still seems to hang after calling the QwenVL node [QwenVL] Node on nvidia_gpu
[QwenVL] Attention backend selected: sdpa
[QwenVL] Loading Qwen3-VL-8B-Instruct-Abliterated (8-bit (Balanced), attn=sdpa)
Loading checkpoint shards: 100%|██████████| 4/4 [00:12<00:00, 3.10s/it]
[QwenVL] torch.compile enabled
@pyeeater283 Disable torch compile in my Qwen node, if you have that enabled the node have to use the graphic card
@huchukato thanks for all your help dude , i did a fresh install of comfy also , resolved all my remaining issues , btw love this model , your work is appreciated
im not sure if its just my computer but i cant get the GGUF models to load on my gpu, they seem to only run on cpu even when the device is set to cuda 0 and i change the layers it still stays in cpu, the cmd says device=cuda though
:\ Don't know
man i don't know what am i doing wrong. i managed to install the qwenvl node but i feel it doesn't work because i don't see it writing any prompt. I tried doing 5 sec, but all i get is my Image fading to a grayscale, nothing else.
which WF are you using, the ones with Qwen GGUF node or with the normal one?
@huchukato i was using one called wan 2.2 i2v svi autoprompt GGUF 1-1, not sure what was the problem. now i'm using one that is called wan2260fps, that i got copying a workflow from a videos metadata. it is working now, but the pictures look blurry. i'm trying to make some anime nsfw videos,but maybe my models are the culprit? i'm kind of confused with the amount of models and wf that exist. the one i'm using is wan22enhancedNSFWSVICamera_nolighting.
But no idea what is svi haha.
Can you recommend me a good workflow and model? i have a 5060ti 16gb VRAM and 32gb or ram
@raidou88 the SVI WG requires the SVI loras and also the Lightx2v loras, its a bit complicated to use, try with the Full-I2V-Autoprompt normal, no GGUF, for the model use the ones I linked inside the WF
Start with the Single Video WF maybe so you will better understand the node https://civitai.com/models/2320999?modelVersionId=2624175 and than go with the long video one https://civitai.com/models/2320999?modelVersionId=2613591
I’m a beginner, but I really love your workflow. I’m using the FP8 model, and QwenVL sometimes causes issues on my setup (and can slow things down).
If possible, could you please make an SVI version without Qwen/autoprompt (or add a simple toggle to disable QwenVL)? That would be hugely appreciated. Thank you!
Try this one, is from the guy that mades the models I use, I will work on a WF without autoprompting these days https://civitai.com/models/2079192?modelVersionId=2668801
using t2v with i2v enabled (if i disable i2v, working good) in autoprompt long video, keep seeing the following error
everything is updated, clip nodes linked (tried both GGUFand safetensors with suggested models)
[QwenVL] Loading GGUF: Huihui-Qwen3-VL-8B-Instruct-abliterated-Q8_0.gguf (device=cuda, gpu_layers=-1, ctx=32768) llama_context: n_ctx_seq (32768) < n_ctx_train (262144) -- the full capacity of the model will not be utilized [QwenVL] Tokens: prompt=800, completion=485, time=22.06s, speed=21.99 tok/s [QwenVL GGUF] Cached new prompt for seed 1989352401: 327c4956... !!! Exception during processing !!! 'str' object has no attribute 'tokenize' Traceback (most recent call last): File "D:\VorteX\comfy\ComfyUI_windows_portable\ComfyUI\execution.py", line 527, 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 "D:\VorteX\comfy\ComfyUI_windows_portable\ComfyUI\execution.py", line 331, 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 "D:\VorteX\comfy\ComfyUI_windows_portable\ComfyUI\execution.py", line 305, in _async_map_node_over_list await process_inputs(input_dict, i) File "D:\VorteX\comfy\ComfyUI_windows_portable\ComfyUI\execution.py", line 293, in process_inputs result = f(**inputs) File "D:\VorteX\comfy\ComfyUI_windows_portable\ComfyUI\nodes.py", line 78, in encode tokens = clip.tokenize(text) ^^^^^^^^^^^^^ AttributeError: 'str' object has no attribute 'tokenize'Delete the Qwen3-VL model in the LLM directory, select the 4B model and download it again, try and let me know
@huchukato absolutely the same. Tried with reinstalled comfyui. Also should mention, inside a Subgraph there are a correct prompt generated. The error is after that generation
!!! Exception during processing !!! 'str' object has no attribute 'tokenize'
Traceback (most recent call last):
File "D:\VorteX\ComfyUI_windows_portable\ComfyUI\execution.py", line 527, 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 "D:\VorteX\ComfyUI_windows_portable\ComfyUI\execution.py", line 331, 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 "D:\VorteX\ComfyUI_windows_portable\ComfyUI\execution.py", line 305, in asyncmap_node_over_list
await process_inputs(input_dict, i)
File "D:\VorteX\ComfyUI_windows_portable\ComfyUI\execution.py", line 293, in process_inputs
result = f(**inputs)
^^^^^^^^^^^
File "D:\VorteX\ComfyUI_windows_portable\ComfyUI\nodes.py", line 78, in encode
tokens = clip.tokenize(text)
^^^^^^^^^^^^^
AttributeError: 'str' object has no attribute 'tokenize'
@vortex28201 I have the same Error. @huchukato on my side it fails everytime on the second batch ("10 sec"), first batch without a problem. Can you please give us a fix for this problem?
@DaDom Hi! I understand you're encountering this error with the CLIP tokenizer. This is actually a common issue that occurs when the CLIP tokenizer object isn't passed correctly to the node.
🎯 Quick fixes to try:
Check your workflow connections: Make sure you're connecting a proper CLIP tokenizer output to the node input, not a text string
Update ComfyUI: Ensure you're using the latest version of ComfyUI (v0.13.0+ recommended)
Verify node setup: Make sure you're using the correct CLIP tokenizer node - try using CLIPVisionModel or CLIPTextEncode nodes instead of passing raw text
Reinstall our custom node:
Delete the ComfyUI-QwenVL-Mod folder from custom_nodes
Restart ComfyUI
Reinstall from the latest release
🔍 If the issue persists:
Check if you're mixing different CLIP model types
Verify your transformers installation is up to date
Try a fresh ComfyUI installation
The code works correctly on our end, so this appears to be a local setup issue. Let me know if you need help with any of these steps!"
@huchukato Can you please specify? Tokenizer node (where do i get it?) to which node? thx in advance..
@DaDom Which workflow are you trying to use?
@huchukato OneClick-I2V-Story and One-Click-T2V-Story. Error happens on both.
@DaDom Hi! Now I see the real issue:
🔍 Error: 'str' object has no attribute 'tokenize'
📍 Location: ComfyUI nodes.py line 78
🎯 Cause: CLIP variable contains text instead of CLIP object
📋 This is a ComfyUI CLIP loading issue, not workflow-related
🔧 Solutions to try:
1. Clear ComfyUI cache: Delete models/clip_vision cache
2. Reinstall CLIP models: Fresh CLIP model download
3. Check QwenVL-Mod version: Update to latest
4. Restart ComfyUI: Clean restart after cache clear
🎯 The subgraph generates correct prompts, but CLIP tokenization fails afterwards
This is a known ComfyUI issue with CLIP model loading. Try the cache clear first!
@huchukato Reinstalled the qwen-mod now over comfyui-manager. Now following error appears:
!!! Exception during processing !!! 'NoneType' object has no attribute 'get_model_object'
Traceback (most recent call last):
File "J:\ComfyUI\ComfyUI_windows_portable\ComfyUI\execution.py", line 530, 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 "J:\ComfyUI\ComfyUI_windows_portable\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 "J:\ComfyUI\ComfyUI_windows_portable\ComfyUI\execution.py", line 308, in asyncmap_node_over_list
await process_inputs(input_dict, i)
File "J:\ComfyUI\ComfyUI_windows_portable\ComfyUI\execution.py", line 296, in process_inputs
result = f(**inputs)
^^^^^^^^^^^
File "J:\ComfyUI\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-WanMoeKSampler\nodes.py", line 148, in sample
model_high_noise = set_shift(model_high_noise, sigma_shift)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "J:\ComfyUI\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-WanMoeKSampler\nodes.py", line 71, in set_shift
model_sampling = model.get_model_object("model_sampling")
^^^^^^^^^^^^^^^^^^^^^^
AttributeError: 'NoneType' object has no attribute 'get_model_object'
I only use the workflow, no change.
@DaDom It doesn't depends on my node or workflow, you downloaded one of the clip models needed by Wan? Are linked in the WF, both normal and NSFW, the NSFW is this one https://huggingface.co/NSFW-API/NSFW-Wan-UMT5-XXL/resolve/main/nsfw_wan_umt5-xxl_fp8_scaled.safetensors the normal UMT5 is in the "Model Manager" in the Comfy Manager
@huchukato I already had all models downloaded for your workflow from other workflows...
@huchukato Just found your error: you connected the positive prompt text encode only input text, but needs input clip too. now it works!!
@huchukato Nevermind, the "get_object" error appears again.....
@DaDom Download the WF again from here the I fixed that link issue
new to all this.. installed the I2v full 1.8 version, says im missing a bunch of nodes and i need to download them..but i cant find any downloads for nodes on this page? where do i get them?
from the ComfyUI Manager https://github.com/Comfy-Org/ComfyUI-Manager
Hey there!! Any chance to integrate CacheDit into the GGUF workflow? I have been using it for a few days and it really increase the speed of generations! https://github.com/Jasonzzt/ComfyUI-CacheDiT
I try it tonight thanks for sharing <3
I was reading how it works and I think it is designed for the models without the Lightx2v LoRaS coz it uses the first 3 steps to warmup
I tryed a lot of times but Comfy says "Failed to import" when I try to load the node
@huchukato I ran into something similar, I asked grok (ai) and it worked after installing some stuff i dont remember haha
First of all, thanks for sharing this great workflow.
I suspect that the auto-prompt generation has built-in NSFW censorship or safety filters enabled. It seems to sanitize my explicit prompts, resulting in safe outputs no matter what I input.
I’ve already tried using NSFW Text Encoders and adjusting settings to bypass this, but nothing seems to work so far.
Any advice would be appreciated!
I'm having the same issue, I downloaded the uncensored version from GitHub in my workflow, but the generated prompt is still censored.
There are no filters in the prompt presets I wrote - in fact, I specifically designed them to emphasize NSFW content inclusion. I've since added additional rules to further improve NSFW prompt adherence. If you're still experiencing this issue, it doesn't come from my node but from the Qwen3-VL model itself. Update my node and let me know, thanks <3
I also update the Qwen3-Vl models, now you will find a josified model in the normal node and 2 new abli gguf models in the gguf node, let me know
Thank you for the update!
I updated the node and tried the Josephized model, but the auto-generated prompt is garbled.
@tamaken0127537 you use my preset prompts? The 3 "Wan" presets I mean, all the videos you seen are generated with that presets
@huchukato I downloaded ComfyUI-QwenVL-Mod and did all the updates. However, the prompts still seem to be unoptimized for NSFW. I'm using Qwen3-VL-4B-Instruct-Abliterated, with the Wan 2.2 I2V preset.
Hello. Thank you so much for sharing your workflow.
Would it be okay to ask two questions?
Q1.
I'm getting the following error when trying to run 10s/15s/20s. Could you help me?
AILab_QwenVL_Advanced
t:1 must be larger than temporal_factor:2
I asked GPT, and they said, "The current incoming frame is 1 (t=1), but the internal video encoder needs at least that many more frames to temporally downsample (temporal_factor=2)."
The QwenVL node has an empty image input and is only connected to the video input (IMAGE batch). The video input is receiving an image batch from ImageScaleBy (id=1260), and the batch is 1 frame long, which is why t=1 is displayed. They also recommend connecting to the "image" port on QwenVL. Is this the correct way to do this?
Q2.
If you run the workflow in the way recommended by GPT, the output video becomes blurry towards the end. Could you please let me know how I can prevent this blurring?
Thank you again for sharing your workflow and for reading this long post. I'd appreciate it if you could reply when you have time.
mmm which version of the WF are you running? Also, update my node coz I did some updates
@huchukato thanks. I'm using OneClick-I2V-Story workflow and it's working well but in this workflow I had to unpack all the subgraph in confyui desktop app.
https://civitai.com/models/2320999?dialog=commentThread&commentId=1116403
I have encounterd same error with this user. please check it out. and thanks for amazing workflow!