A few days ago, I shared many cases of FLUX Kontext Although I roughly explained how to use it, many people still don't know how This time, I just discovered that Liblib has launched Kontext, which doesn't require running Comfyui locally and can be processed online I'll teach everyone step by step And share some common Comfyui workflows

Finally, the resolution of the images generated by Kontext is not very high, so you may need to enlarge the images in many cases. There are many other infrastructures in liblib's Comfyui, so Master Zang found an image enlargement process and integrated it into it. Just drag the "FLUX Kontext three-image plus enlargement workflow" file to the comfyui interface. The parameters you need to pay attention to are still the same, and you don't need to worry about the enlargement part at all, it's all automatic.

And I also added a FLUX Lora that I trained, which can remove the FLUX very well. You can see that the character's skin texture, clothing material and picture color are much better. Of course, this super-resolution will definitely have some impact on consistency, so if you care about it, you can use it.