I believe this model to be very good at photorealism although it does have some artistic capabilities as well. Here are a few tricks to using the SDXL versions of GonzaLomo to its fullest potential:
Clip Skip is your friend. I recommend trying a prompt with both 1 and 2 values. 1 can make images that are more creative, artistic, retro or grungy whereas 2 is better at photorealism and the, um, physical attractiveness of the subject(s).
Generate at 1024x1536 or higher using the PatchModelAddDownscale node in ComfyUI. This will create a much more detailed image that can then be upscaled further, resulting in incredible photorealistic detail. Using that node, I can do an initial generation as high as 1280x1920.
You can play around with different sampler settings but I find that the absolute best results are with LCM/Karras, 12 steps and CFG 1.0.
Selective usage of the Self Attention Guidance node can sometimes give amazing results. I always have it in the pipeline and usually bypassed but turn it on when I think an image has potential but just needs a little more oomph like color saturation or contrast.
I get really interesting/creative results by generating with the V1.1 Semi-Real model and then resampling that image with the V2.0 Unity model at around 0.60 denoise. You really need to play with the denoise to find the right value that tips the image from cartoonish to photorealism but you'll be amazed at what's possible when you do. Note that V1.1 Semi-Real requires Clip Skip 2+ and will not work with 1 at all.
And lastly, always be upscaling! More pixels equals more detail equals more photorealism.
Note: I only use ComfyUI so I apologize in advance if you're having trouble with my model in some other tool and I can't help.
Current Models
V2.0 Unity
Essentially, this is V1.1 Real with a bit of Siren Forge, Cyberrealistic and Lustify added to the mix. The base is still Gonzales and LomoXL.
I called this version Unity because I originally wanted to have separate models for photorealism and artistic styles. However, I realized that I could get both in one model and created V1.1 Artsy for that purpose but it was a failure. This is my second attempt at that combination and I think/hope it's a success.
At the very least, I don't want to degrade the photorealism capabilities and just want to add more surreal, artistic or fantastical styles as an option.
The recommended settings haven't changed since V1.1 Real:
Sampler: LCM (also dpmpp2a and ddim can provide interesting results)
Scheduler: Karras, Exponential or Beta
Steps: 8-12
CFG: 1.0-1.3
Clip Skip: 1-5
I normally keep my settings at LCM/Karras with a CFG of 1.0 but I tend to vary the clip skip between 1 and 5 because those can change the image pretty drastically. Sometimes an image at clip skip 2 will be terrible but might become amazing with clip skip 1, and vice-versa.
V1.1 Semi-Real
I'm doing a little experimenting with non-photorealistic models. This is a merge of GonzaLomo v1.1 Real and ToonComix which is a toon model that does a really good job with detailed backgrounds. The whole Comix family of models is very good so check them out.
Sampler: LCM
Scheduler: Karras, Exponential or Beta
Steps: 8-12
CFG: 1.0-1.3
Clip Skip: 2-5 (1 is not supported)
Retired Models
V1.1 Real
This is a simple 60/40 merge between two of my favorite photorealistic models - Gonzales-NSFW-PonyV1/V2-DMD v2.0 and LomoXL (with a little bit of Lustify thrown in). Many, many thanks to the makers of those excellent models.
Gonzales is a Pony model so GonzaLomo inherits all of those capabilties. LomoXL has a very strong analog feel to it which helps give an edge to Gonzales' "pony-ness".
Sampler: LCM
Scheduler: Karras, Exponential or Beta
Steps: 8-12
CFG: 1.0-1.3
Clip Skip: 1 allows greater analog effect but lessens NSFW capabilities
Clip Skip: 2-5 lessens/removes the analog effect but allows more NSFW
Try this workflow to get started, if you need it.
To be clear, I created this merge for my own use cases. It's certainly not better than Gonzales for most things. I'm just hoping to inject Gonzales with some of the gritty, analog feel that I was looking for in some cases. LomoXL also allows for a wider variation of faces, especially with clip skip 1.
Enjoy!
P.S. This is my first model upload so I would appreciate any constructive feedback or comments.
V1.1 Artsy alpha (Not very good imo, use V2.0 Unity instead)
This was an attempted merge between V1.1 Real and a few art-focused models. The main purpose of this model is to be able to produce artistic styles but it has the tendency to try to force photorealism when prompted clearly for an artistic style. This model has been abandoned in favor of V2.0 Unity which does a much better job of combining the photorealism of V1.1 Real with artistic styles.
Description
GonzaLomo v6.0 without the integrated DMD2 LoRA.
Recommended settings:
Sampler: dpmpp_3m_sde or dpmpp_2m_sde
Scheduler: Exponential or Karras
Steps: 20-30
CFG: 3.0-5.0
Clip Skip: 1-5 (I like 2 best)FAQ
Comments (37)
What is the difference between a model with and without dmd?
DMD is a lora that speeds up the image generation but also can help create very high quality details. There's a tradeoff though and some people don't like it because some texture detail can be lost.
@GBRX , Cheers! Another top quality Checkpoint
The models and workflows are top notch. Super useful. Thank you! My sense is that the sdxl models used as refiners tend to add in tan lines often, even when set as negative prompt. Any ideas in preventing this?
Thanks for the kind words! I wish I had a good solution for the tan lines. What works best for me is 'pale skin'. Otherwise, you can use one of the non-DMD models and add 'tan lines' to the negative prompt.
V6 so good thank you @GBRX <3
idk what i'm doing, but unity 4 is incredible, great work
what are the differences between all these versions? is V6 updated V4 unity? and what is the diff between dmd/non dmd? looks great cant wait to try!
V6 is the best of my photorealistic models, a spiritual successor to V4 but not a direct update. "DMD" refers to a lora that gets applied to an SDXL checkpoint to make it much faster and also add some additional training to the checkpoint. Some people like DMD and some don't but I've got both versions to try if you're curious.
@GBRX ty
@GBRX Much appreciated!
@GBRX is the lora merged into it ,or we gotta apply it still?
@Meower2024 If the model is named 'XL DMD' then it already has the lora applied to it. If it says 'Xl Non-DMD' then it doesn't.
Hi @GBRX
I've been really enjoying your models, especially the Gonzalos model with its efficient 8-step generation. However, I'm working with several LoRAs and IPAdapter features that are only compatible with SD1.5.I checked your collection but didn't find any SD1.5 models. Would you be able to:Recommend a similar SD1.5 checkpoint that matches the quality and speed of your Gonzalos model, Or consider converting the Gonzalos model to SD1.5?I've tried other checkpoints but haven't been able to achieve the same results I get with your models. Any suggestions would be greatly appreciated!Thanks!
Thanks, I've considered doing a SD 1.5 model but it's always been at the bottom of the todo list. As far as recommendations go, you can probably just sort the SD 1.5 checkpoints by Highest Rated and check those out. I'd probably go with Analog Madness, LazyMix, EpicPhotoGasm, Realistic Vision or Absolute Reality.
Is this an image or a video model? If it is an image model, then showcase images and not videos. We know that you can do videos out of the images but that's for a different model not this one.
I don't really understand this comment. I think I'm showcasing plenty of images with all of these models that a few videos are not going to matter. I disagree that I shouldn't be able to post videos made from the images that the model creates because the quality of an I2V video is heavily dependent on the quality of the start image.
If we're talking about people then you can use a phrase like full body. Also, the more you describe in the prompt then the more you will see in the image. If you say wearing cowboy boots and a a cowboy hat then you will also get more of a full body image. Usually it's better to just be descriptive about what you want to see rather than try to use keywords.
Your models and workflows are brilliant, thank you so much. Can't believe how well they work with my LORA's and the results are just beautiful. Great work.
That's great, thanks for the feedback!
when i load the workflow its missing SDXLEmptyLatentSizePicker+
any ideas?
You just need to do Install Missing Custom Nodes in the ComfyUI Manager.
Works great except for i2i. i2i gives me like strange textures artifacts, the lower the denoise (0.4-0.5) the stronger the artifact, at higher value it disapears gradualy but the source image isn't respected at all anymore. Is it expected or am I missing something?
There's a lot of models on this page. Which one are you referring to?
@GBRX oh sorry, gonzalomoXLFluxPony_v60PhotoXLDMD is the one I'm referring to
@kito12923 Do you have an image you can point to as an example? That model is one of the primary models I use for I2I refinement in my workflows and I haven't seen the issues you're describing.
Gostei baste, muito bom
what does the DMD mean?
It refers to the DMD2 lora that speeds up generation and adds detail.
Hi, where can I find ultraSense10-50_ultraFinetunev4-50_turbo_lenovo-lora-50_00001.safetensorsю? The model is in your work, but I can't find it.
You can disregard that name. I accidentally left the filename in the Save Checkpoint node on an older test merge. The workflow you're referring to is using the GonzaLomo Flux D v2.0 model, https://civitai.com/models/1513492?modelVersionId=2415956
I have just started using V6.0 Photo XL DMD and pair it with either DPM++ SDE or DPM++ 2M SDE and karras with some fantastic results. I find the magic sauce is around CFG 5-8 with steps 25 (only for the higher CFG) and 30 or 35 for the others.
Thanks!
How on Earth do you do clip skip for SDXL and ComfyUI? Am I missing something?
CLIP set last layer?
If you look at my SDXL workflow you'll see how it's done, https://civitai.com/models/1513601/gonzalomo-sdxl-workflows
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.