Z-Image is here, and it's pretty damn great. Let's make it more NSFW!
Recommended Settings
Steps: 12
CFG: 1.0
Sampler: dpmpp_sde
Scheduler: simple
FP8 Download
For some reason, when you upload multiple formats (BF16 + FP8) on one model, which they now support, they hide the FP8 under the BF16 model group in the side menu...
ComfyUI
Models should be placed in diffusion_models, and uses the official workflow with the Load Diffusion Model loader, and loading the VAE/CLIP models separately. Models named "_AIO" should go in the checkpoints folder, and uses the normal checkpoint loaders.
Description
FAQ
Comments (6)
These are some of my favorites, thanks for the awesome checkpoints. An issue is v9 mangles many of the more complex prompts compared to v8 side-by-side, just so you're aware.
Could you give a side-by-side example? Reference of a mangled prompt.
@6tZ Unfortunately I already deleted v9 and kept v8, it was definitely mangling a lot of the more detailed NSFW prompts that 8 could handle. I don't recall exactly what prompts, but quite a few which weren't close ups. Most people probably do close up by I prefer wide shot detailed scenes and that's where v9 seemed to have gotten worse than v8.
@ferrrett33 Got it, thanks for the suggestions, will add it to the list. The next version is already nearly done, so it'll be a while before I get onto this.
@ferrrett33 What are you talking about? I just compared v8 and v9.2, and I'll say that this is heaven and earth, there are a lot of details in the 9.2 version, promta listens better than all the others, I was surprised myself
Kudos are due ... I'm glad you've been keeping up on creating new versions of this. My personal favorites as of now are 5.0, 8.2, and 9.0. It's great being able to choose different styles depending on what look I'm going for with a prompt, and the diversity in the newer checkpoints has been quite welcome.