Flux2 Klein Multi-view Character Generation (6G VRAM can run it!)

⚠️⚠️⚠️You need to install this plugin manually, otherwise an error will occur when loading the workflow.【ComfyUI-EditUtils】（https://github.com/lrzjason/ComfyUI-EditUtils）⚠️⚠️⚠️

More workflow parameter settings, please refer to my settings on rh:

https://www.runninghub.ai/post/2022312745244299266/?inviteCode=rh-v1315

Click the link to download the workflow. After registering and binding my invitation code: rh-v1315, you can get an extra 1000 RH coins. You can also get 100 points by logging in daily. It can be used to generate images or videos.

I have published a video introduction:

v2:

I tried it, and it works fine with 6GB of local video memory.

My local configuration is: RTX 3060 laptop version (6GB VRAM) + 16GB RAM. The entire process took about 6 and a half minutes.

However, the model needs to be changed to 4b. After changing to the 4B model, the background log will report an error indicating that the Lora is mismatched (because both of these Lora are derived from the 9B model. I found a 4b model for face swapping, but there is no 4b version of consistency Lora yet). But don't worry about it, it can still be generated normally.