Realism......
I design the most professional workflows out there. I have better options, better samplers, better detailers, better prompt assist, better controlnet, and basiclly better everything, and I give them away for free. I'm not going to charge you $30 for a crappy workflow with a renamed Clownshark sampler, a horribly vibecoded detailer, and a bad Lut. I give you an almost infintite amount of adjustments and leave you notes on how to do it.
If here is a feature you want, I'll add it. If you have questions, I answer them. I also have several other realism workflows. Just don't waste your money.
All of the images I posted to V7 were zero LoRAs and basic settings. You can make any adjustment you want. I give you the tools to do it.
Note: If you have ANY issues with nodes not downloading, read the notes or reach out. There's nothing that special about any of them that aren't core modules.
⛔⚠️🛑✋ Read the notes completly before using. Most common install and node problems are listed in the directions
也有中文说明
This is an advanced workflow. I encourage you to take the time to learn it. If you feel overwhelmed, you can start here:
https://civarchive.com/models/2348914/zimage-base-and-turbo-essentials-workflow
A complete user and troubleshooting guide inclusing a list of common error messages can be found here https://civarchive.com/articles/24678/how-to-use-my-workflows-a-comprehensive-breakdown
All in 1 workflow:
Controlnet
Checkpoint option for merged models
I2I
Prompt generation
Prompt enhancement
Lora Loader
Flowmatch Sampler (Rez4Lyf)
Draft mode for making corrections without having to rerun the workflow.
Seed variance enhancer
Ultimate Upscaler for prescaling and hi res fix
Detailing Suite for face, hands, eyes, and expressions
NSFW detailers
Seed VR2 Upscaler to get 4k on low VRAM
Massive Post Production Suite
Smart Noise scrubber
Full tutorial available.
也有中文说明
Check out my other models:
My NSFW version with Anantomy Correctors:https://civarchive.com/models/2270894/zit-nsfw-anatomy-corrector
For ultimate Anime detail and control: https://civarchive.com/models/2220766/zimage-ultra-anime-detail-workflow-get-the-most-out-of-your-generations
For Realism and variety over the base model: https://civarchive.com/models/2231181/z-image-ultra-real-workflow
Instagram: https://www.instagram.com/synth.studio.models/
Buy me a☕ https://ko-fi.com/lonecatone
This represents hundreds of hours of work. If you enjoy it, please 👍like, 💬 comment , and feel free to ⚡tip 😉
Description
2 months worth of upgrades rolled into one:
swapped out the primary sampler for a dual pass for better Controlnet images
Massive controlnet overhaul
Overhauled the Image and prompt generation area
Upgrade to detailers
Upgrade to Ultimate Upscale
Added inpainting suite
upgraded Post production Suite
Modified math nodes to better work with Linux.
FAQ
Comments (22)
Really been enjoying your workflows! Thanks for the Linux boost =)
Hopefully it helps. If not, tell me exactly what wentr wrong. I rpleced all the math expression nodes.
Let me know about how it saves (\ vs. \\) so I can adjust
@lonecatone23 Oh, on that, I still wound up changing the delimiter. I think for Linux its / instead of \ . Otherwise the folder created was also including the stuff after the metadata node. But I also like to create a subfolder in a subfolder lol, ie: output/zimage/%date/ then stuff. easier for me to keep it kinda organized. I'm also still kinda new with linux, so that's just my experience! I've not had any other issues besides the sliders from the mx node pack not showing up properly, but that may be another issue all together! i've not troubleshoot that one, probably conflicting with other popular node packs from what i've gathered ><
@Neriva Let me know. There are alternates, bur the MX node pack is unfortunatly the power player of the game. Let me know whet you replaced them with
@lonecatone23 I replaced some of the float ones with one i had from a pack called levelpixel. cant remember why i have that pack but i do! Weirdly enough, it runs just fine if i dont replace them, I just cant adjust the values. The nodes dont show the slider part. Probably, some update or pack caused it, cause they did used to work about a month or so ago. I just havent gone through one by one and tried to figure it out lol.
@Neriva technically you can right click and adjust teh values in the properties menu.
@lonecatone23 Oh! awesome! I didn't even think of that lol. Thanks! no need to replace really, then.
Hi, I am using your workflow : zImageBaseTurboProGrade_v70MassiveUpgrade
It passed Florence2Run, Background Remover but threw error in "Samplers" group (Dual Pass), there we have 2 Ksamplers shown like collapsed mini node, error happened in 1st kSampler.
# ComfyUI Error Report ## Error Details - Node ID: 1786 - Node Type: KSamplerAdvanced - Exception Type: RuntimeError - Exception Message: RuntimeError: Given normalized_shape=[2560], expected input with shape [*2560], but got input of size[1, 298, 4096] ## Stack Trace ```
Can you please tell what I am doing incorrect . Also I have gone through your tutorial page : https://civitai.com/articles/24678/how-to-use-my-workflows-a-comprehensive-breakdown-and-troubleshooting-guide
But it is bit overwhelming. I want to learn this advanced workflow of yours like a story , understanding the intricate workings so that i can play with the settings myself.
that is always a mismatched model and clip. it will fault at the ksampler.
Make sure you are using teh right model and clip combination. Message me with your stacktrace
Yes, you were absolutely correct, it was the clip loader only , i loaded the wrong one. Now i am running into new issue :: gettting OOM in Qwen Prompt enhancer ::
[WARNING] unet unexpected: ['norm_final.weight']
[QwenVL] Node on nvidia_gpu
[QwenVL] BitsAndBytes quantization detected - forcing SDPA attention
[QwenVL] Attention backend selected: sdpa
[QwenVL] Loading Qwen3-VL-4B-Instruct (4-bit (VRAM-friendly), attn=sdpa)
[ERROR] loss is part of Qwen3VLCausalLMOutputWithPast.__init__'s signature, but not documented. Make sure to add it to the docstring of the function in D:\ImageGenProj\comfyUI setup\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\models\qwen3_vl\modeling_qwen3_vl.py.
[ERROR] logits is part of Qwen3VLCausalLMOutputWithPast.__init__'s signature, but not documented. Make sure to add it to the docstring of the function in D:\ImageGenProj\comfyUI setup\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\models\qwen3_vl\modeling_qwen3_vl.py.
Loading weights: 30%|██████████████████▉ | 215/713 [00:19<00:58, 8.52it/s]Windows fatal exception: access violation
Thread 0x00003bc0 (most recent call first):
File "threading.py", line 363 in wait
File "threading.py", line 660 in wait
File "D:\ImageGenProj\comfyUI setup\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\python_embeded\Lib\site-packages\tqdm\_monitor.py", line 60 in run
File "threading.py", line 1044 in bootstrapinner
File "threading.py", line 1015 in _bootstrap
I am on 1660ti 6GB vram.
@john20200501 so there is a 2b version, but realistically, you'll never be able to run it on a system that small in any reasonable fashion.
I have a beats (rtx5090 and 128gb RAM). I never use it. I leave it in as numerous peopel have asked fo it. You can delete it in its entirety. Just make sure that the promp text string remains intact
@lonecatone23 Thanks a lot man! Ok a few queries here, sorry for my dumbness, i am learning things.
1) Inpaint Group -> For inpainting what i normally did in past is to load an image > Open in mask editor then manually paint the area i want to change . But here, i do not see any such system in your inpaint group , Will just the prompt be enough to put on any outfit?
2) Where do i add my "face lora" which is trained on my consistent character faces. There multiple lora loader nodes across the groups. I guess its the "User Settings" group's lora loader, i need to put it in? And what about trigger word? where can i add that.
3) In the post processing group , there are multiple switches to enable features like aberration, arcane bloom, smart denoise etc , so this is just a hit and trial, right? like which combination of these features will work for me, and product better images for my usecase? Actually i generated an image that passed the post production group, yesterday, but image got converted to be black and white. I do not know what feature(s) converted it to black and white.
Btw I must say, the workflow is phenomenal. You have already provided heavy abstraction on the node connections and all, making it easier to atleast get a high level understanding.
@john20200501
yes on inpaint,
load your LoRA in teh main power LoRA loader. Also in the detailers if you want to.
NMost of teh switches are for art. Te basics:
Image adjust, Lens FX, Lut, and optical realism if you wnt to dail it in. The rest just do cool stuff. Fix the seed and mess with it.
I do my best to lay it out so that people can follow along and make it so you can take an entire module out and paste it into yours.
Hi! I’m trying to set up your Lonecat’s ZIT Realism 7.0 workflow, but I’m missing one model file: zImageZITTA86Allround_v01FP8.safetensors I couldn’t find it anywhere online. Could you please let me know where I can download it, or if it has been renamed or replaced by another model? Thank you very much for your help!
I think you messaged me @AllanTassie. Please hook them up
@lonecatone23 He messaged me now, I gave him the link
Hi @AllanTassie , I have the same issue. Could you please send me the link for zImageZITTA86Allround_v01FP8.safetensors ? Thanks
@tonytk Pretty sure this is the one you want... https://civitai.red/models/2344548/z-image-zit-ta86-allround-aio?modelVersionId=2660158
@tonytk there is a FP16 version too
@fedechkoyursi114 there is! The link leads to it as well!
@tonytk looks like @fedechkoyursi114 has already sent it to you :-)
@AllanTassie Lol, no clue what I tagged you in Halex's checkpoint. I guess I need sleep.
Thanks tho.