It is not a very complex or difficult workflow.
I am publishing this workflow to answer questions like "your image has a lot of detail, how did you do it", "can you share a workflow?".
Of course, it works well on other models, but I would like you to use my FRT models and give feedback if possible.
This workflow is in order:
Image generation with Lightning model,
Upscale with SD1.5,
there are also elaboration and correction processes with SD1.5.
SD1.5: https://civarchive.com/models/286334/frt-mixmodel
Lightning: https://civarchive.com/models/433850/frt-mixmodel-sdxl
Description
FAQ
Comments (5)
This is a perfect workflow! Thanks! great job :)
Thank you ;)
Hi, import via manager in comfyui: Image2Prompt failled. Nothing work.
it's now year gone you didn't get reply and I am here for a same error maybe I need to delete this workflow lol
Almost 100GB downloading to activate a plugin?
Step 2: Download the Model The model will be automatically downloaded the first time it is run. If it does not download normally, for the plugin to function properly, you need to download the necessary models.
This plugin utilizes the
vikhyatk/moondream1
vikhyatk/moondream2
unum-cloud/uform-gen2-qwen-500m and internlm/internlm-xcomposer2-vl-7b
models from Hugging Face. Make sure to download these models into the plugin's ComfyUI/models/image2text directories, respectively.
Use the following links for downloading:
Download moondream1 Model
Download moondream2 Model
Download internlm-xcomposer2-vl-7b Model Download uform-gen2-qwen-500m Model Download Qwen-1_8B-Stable-Diffusion-Prompt Download deepseek-vl-1.3b-chat
Download deepseek-ai/deepseek-vl-7b-chat
??