Innovate Futures @ Benji
patreon
ComfyUI Qwen 3 VL Create Powerful Prompts For Image And Video Generation
🕑 Added 2025-10-21 15:06:01 +0000 UTC
Video : https://youtu.be/3j9c_-mRKfg
In this video, you’ll learn how to use ComfyUI Qwen 3 VL—a powerful vision language model—directly inside ComfyUI to generate detailed text prompts from images or videos, and then use those prompts to create new AI-generated content. We walk through both image and video workflows, showing how Qwen VL can analyze visual input and produce rich, time-coded descriptions that feed into diffusion models like WAN 2.2 or SDXL. Whether you're refining images with multi-stage sampling, applying LoRAs for style control, or generating synchronized video narratives, this tutorial gives you a practical, local, and customizable pipeline. This content is perfect for AI artists, ComfyUI users, and creators who want to move beyond basic prompting and explore dynamic, vision-driven generation. It matters because it bridges advanced multimodal AI with real-world creative workflows—no cloud APIs, no subscriptions, just local control and creative freedom.
Resources:
Qwen3-VL-4B-Instruct
https://huggingface.co/Qwen/Qwen3-VL-4B-Instruct
Qwen3-VL-4B-Instruct-FP8
https://huggingface.co/Qwen/Qwen3-VL-4B-Instruct-FP8
Qwen3-VL-8B-Instruct
https://huggingface.co/Qwen/Qwen3-VL-8B-Instruct
ComfyUI-QwenVL
https://github.com/1038lab/ComfyUI-QwenVL
Tutorial Example Workflows :
Comments
Steve Lam
I got this error but i do not know what happen... Anyone can help me up? Prompt outputs failed validation: AILab_QwenVL_Advanced: - Value not in list: preset_prompt: 'Describe this image in detail.' not in ['Prompt Style - Tags', 'Prompt Style - Simple', 'Prompt Style - Detailed', 'Prompt Style - Extreme Detailed', 'Prompt Style - Cinematic', 'Creative - Detailed Analysis', 'Creative - Summarize Video', 'Creative - Short Story']
Jason Walsh
Hey Max- it's telling you what you need to do. Upgrade Transformers. If you aren't familiar with using command prompt but you'd still like to do that, go to ChatGPT, tell it you aren't sure how to use command prompt, but ComfyUI gave you that error and you need to know how to fix it. It will take you step by step through the process so you can do it properly. Your Qwen V3 is too new for the current version of Transformers that you have. It needs to be upgraded.
Max Schreck
I am getting this error. What exactly do I need to do? I am quite new to this. ERROR: The checkpoint you are trying to load has model type `qwen3_vl` but Transformers does not recognize this architecture. This could be because of an issue with the checkpoint, or because your version of Transformers is out of date. You can update Transformers with the command `pip install --upgrade transformers`. If this does not work, and the checkpoint is very new, then there may not be a release version that supports this model yet. In this case, you can get the most up-to-date code by installing Transformers from source with the command `pip install git+https://github.com/huggingface/transformers.git`
