[Bug]: Flat 2d Art being ignored

[llagerlof]

Checklist

• The issue has not been resolved by following the troubleshooting guide
• The issue exists on a clean installation of Fooocus
• The issue exists in the current version of Fooocus
• The issue has not been reported before recently
• The issue has been reported before but has not been fixed yet

What happened?

Style "Flat 2d Art" being ignored, mostly. Sometimes it follows the style, sometimes, not.

Steps to reproduce the problem

1. Write a prompt. I tried "dog walking, german shepherd breed" without quotes.
2. Tick the "Advanced" checkbox.
3. Select only the style "Flat 2d Art".
4. Sometimes it follows the style, sometimes ignore it completely.

What should have happened?

The result should be something like this (one of the times that worked):

What browsers do you use to access Fooocus?

Mozilla Firefox, Google Chrome

Where are you running Fooocus?

Locally

What operating system are you using?

Windows 11

Console logs

c:\Fooocus_win64_2-1-791>run.bat

c:\Fooocus_win64_2-1-791>.\python_embeded\python.exe -s Fooocus\entry_with_update.py --always-download-new-model
Fast-forward merge
Update succeeded.
[System ARGV] ['Fooocus\\entry_with_update.py', '--always-download-new-model']
Python 3.10.9 (tags/v3.10.9:1dd9be6, Dec  6 2022, 20:01:21) [MSC v.1934 64 bit (AMD64)]
Fooocus version: 2.3.1
[Cleanup] Attempting to delete content of temp dir C:\Users\lawrence\AppData\Local\Temp\fooocus
[Cleanup] Cleanup successful
Total VRAM 12282 MB, total RAM 32671 MB
Set vram state to: NORMAL_VRAM
Always offload VRAM
Device: cuda:0 NVIDIA GeForce RTX 4070 : native
VAE dtype: torch.bfloat16
Using pytorch cross attention
Refiner unloaded.
Running on local URL:  http://127.0.0.1:7865

To create a public link, set `share=True` in `launch()`.
IMPORTANT: You are using gradio version 3.41.2, however version 4.29.0 is available, please upgrade.
--------
model_type EPS
UNet ADM Dimension 2816
Using pytorch attention in VAE
Working with z of shape (1, 4, 32, 32) = 4096 dimensions.
Using pytorch attention in VAE
extra {'cond_stage_model.clip_g.transformer.text_model.embeddings.position_ids', 'cond_stage_model.clip_l.text_projection', 'cond_stage_model.clip_l.logit_scale'}
Base model loaded: C:\Fooocus_win64_2-1-791\Fooocus\models\checkpoints\juggernautXL_v8Rundiffusion.safetensors
Request to load LoRAs [['sd_xl_offset_example-lora_1.0.safetensors', 0.1], ['None', 1.0], ['None', 1.0], ['None', 1.0],['None', 1.0]] for model [C:\Fooocus_win64_2-1-791\Fooocus\models\checkpoints\juggernautXL_v8Rundiffusion.safetensors].
Loaded LoRA [C:\Fooocus_win64_2-1-791\Fooocus\models\loras\sd_xl_offset_example-lora_1.0.safetensors] for UNet [C:\Fooocus_win64_2-1-791\Fooocus\models\checkpoints\juggernautXL_v8Rundiffusion.safetensors] with 788 keys at weight 0.1.
Fooocus V2 Expansion: Vocab with 642 words.
Fooocus Expansion engine loaded for cuda:0, use_fp16 = True.
Requested to load SDXLClipModel
Requested to load GPT2LMHeadModel
Loading 2 new models
[Fooocus Model Management] Moving model(s) has taken 0.59 seconds
Started worker with PID 25056
App started successful. Use the app with http://127.0.0.1:7865/ or 127.0.0.1:7865
[Parameters] Adaptive CFG = 7
[Parameters] Sharpness = 2
[Parameters] ControlNet Softness = 0.25
[Parameters] ADM Scale = 1.5 : 0.8 : 0.3
[Parameters] CFG = 4.0
[Parameters] Seed = 7760145174860190951
[Parameters] Sampler = dpmpp_2m_sde_gpu - karras
[Parameters] Steps = 30 - 15
[Fooocus] Initializing ...
[Fooocus] Loading models ...
Refiner unloaded.
[Fooocus] Processing prompts ...
[Fooocus] Preparing Fooocus text #1 ...
[Prompt Expansion] blue bird standing still, deep focus, highly detailed, sharp focused, cinematic, ambient light, dynamic background, elegant, intricate, designed, new classic, epic, great composition, dramatic, aesthetic, confident, unique, creative, attractive, positive, cute, artistic, beautiful, radiant, loving, pretty, color, inspired, vibrant, shiny,bright, illuminated, brilliant
[Fooocus] Preparing Fooocus text #2 ...
[Prompt Expansion] blue bird standing still, highly detailed, sharp focus, cinematic, background, colorful, very strong, open epic artistic, excellent composition, dramatic ambient light, dynamic, atmosphere, rich vivid colors, beautiful grand intricate detail, creative, positive imaginative, glowing, pretty, fine balanced, confident, unique, awesome, inspired, color, illuminated, complex, iconic, shiny, amazing
[Fooocus] Encoding positive #1 ...
[Fooocus Model Management] Moving model(s) has taken 0.16 seconds
[Fooocus] Encoding positive #2 ...
[Fooocus] Encoding negative #1 ...
[Fooocus] Encoding negative #2 ...
[Parameters] Denoising Strength = 1.0
[Parameters] Initial Latent shape: Image Space (896, 1152)
Preparation time: 5.43 seconds
[Sampler] refiner_swap_method = joint
[Sampler] sigma_min = 0.0291671771556139, sigma_max = 14.614643096923828
Requested to load SDXL
Loading 1 new model
[Fooocus Model Management] Moving model(s) has taken 2.20 seconds
100%|██████████████████████████████████████████████████████████████████████████████████| 30/30 [00:13<00:00,  2.26it/s]
Requested to load AutoencoderKL
Loading 1 new model
[Fooocus Model Management] Moving model(s) has taken 0.13 seconds
Image generated with private log at: C:\Fooocus_win64_2-1-791\Fooocus\outputs\2024-05-14\log.html
Generating and saving time: 16.73 seconds
[Sampler] refiner_swap_method = joint
[Sampler] sigma_min = 0.0291671771556139, sigma_max = 14.614643096923828
Requested to load SDXL
Loading 1 new model
[Fooocus Model Management] Moving model(s) has taken 1.09 seconds
100%|██████████████████████████████████████████████████████████████████████████████████| 30/30 [00:10<00:00,  2.80it/s]
Requested to load AutoencoderKL
Loading 1 new model
[Fooocus Model Management] Moving model(s) has taken 0.13 seconds
Image generated with private log at: C:\Fooocus_win64_2-1-791\Fooocus\outputs\2024-05-14\log.html
Generating and saving time: 12.61 seconds
Requested to load SDXLClipModel
Requested to load GPT2LMHeadModel
Loading 2 new models
Total time: 34.82 seconds
[Fooocus Model Management] Moving model(s) has taken 0.42 seconds
[Parameters] Adaptive CFG = 7
[Parameters] Sharpness = 2
[Parameters] ControlNet Softness = 0.25
[Parameters] ADM Scale = 1.5 : 0.8 : 0.3
[Parameters] CFG = 4.0
[Parameters] Seed = 5694575736345519726
[Parameters] Sampler = dpmpp_2m_sde_gpu - karras
[Parameters] Steps = 30 - 15
[Fooocus] Initializing ...
[Fooocus] Loading models ...
Refiner unloaded.
[Fooocus] Processing prompts ...
[Fooocus] Preparing Fooocus text #1 ...
[Prompt Expansion] yellow and red bird standing still, highly detailed, sharp focus, cinematic, composition, colorful, dramatic, scenic, elegant, professional, vivid, very structured, color, artistic,, open background, discovered, extremely beautiful, creative, epic, surreal, inspiring, awesome, cool, inspired, pretty, amazing, attractive, best, clear, perfect, tailored, cute
[Fooocus] Preparing Fooocus text #2 ...
[Prompt Expansion] yellow and red bird standing still, highly detailed, sharp focus, elegant, intricate, cinematic, complex, directed, modern, futuristic, inspired, fancy, illustrious, fine classic, ambient, shiny, bright, color, extremely, very coherent, colors, epic, romantic, artistic, new, sleek, amazing, perfect, awesome, cute, gorgeous, creative, pure
[Fooocus] Encoding positive #1 ...
[Fooocus] Encoding positive #2 ...
[Fooocus] Encoding negative #1 ...
[Fooocus] Encoding negative #2 ...
[Parameters] Denoising Strength = 1.0
[Parameters] Initial Latent shape: Image Space (896, 1152)
Preparation time: 2.69 seconds
[Sampler] refiner_swap_method = joint
[Sampler] sigma_min = 0.0291671771556139, sigma_max = 14.614643096923828
Requested to load SDXL
Loading 1 new model
[Fooocus Model Management] Moving model(s) has taken 1.65 seconds
100%|██████████████████████████████████████████████████████████████████████████████████| 30/30 [00:10<00:00,  2.99it/s]
Requested to load AutoencoderKL
Loading 1 new model
[Fooocus Model Management] Moving model(s) has taken 0.14 seconds
Image generated with private log at: C:\Fooocus_win64_2-1-791\Fooocus\outputs\2024-05-14\log.html
Generating and saving time: 12.54 seconds
[Sampler] refiner_swap_method = joint
[Sampler] sigma_min = 0.0291671771556139, sigma_max = 14.614643096923828
Requested to load SDXL
Loading 1 new model
[Fooocus Model Management] Moving model(s) has taken 1.08 seconds
100%|██████████████████████████████████████████████████████████████████████████████████| 30/30 [00:10<00:00,  2.98it/s]
Requested to load AutoencoderKL
Loading 1 new model
[Fooocus Model Management] Moving model(s) has taken 0.14 seconds
Image generated with private log at: C:\Fooocus_win64_2-1-791\Fooocus\outputs\2024-05-14\log.html
Generating and saving time: 12.01 seconds
Requested to load SDXLClipModel
Requested to load GPT2LMHeadModel
Loading 2 new models
[Fooocus Model Management] Moving model(s) has taken 0.39 seconds
Total time: 27.85 seconds
[Parameters] Adaptive CFG = 7
[Parameters] Sharpness = 2
[Parameters] ControlNet Softness = 0.25
[Parameters] ADM Scale = 1.5 : 0.8 : 0.3
[Parameters] CFG = 4.0
[Parameters] Seed = 5184485171165062298
[Parameters] Sampler = dpmpp_2m_sde_gpu - karras
[Parameters] Steps = 30 - 15
[Fooocus] Initializing ...
[Fooocus] Loading models ...
Refiner unloaded.
[Fooocus] Processing prompts ...
[Fooocus] Encoding positive #1 ...
[Fooocus Model Management] Moving model(s) has taken 0.16 seconds
[Fooocus] Encoding positive #2 ...
[Fooocus] Encoding negative #1 ...
[Fooocus] Encoding negative #2 ...
[Parameters] Denoising Strength = 1.0
[Parameters] Initial Latent shape: Image Space (896, 1152)
Preparation time: 0.62 seconds
[Sampler] refiner_swap_method = joint
[Sampler] sigma_min = 0.0291671771556139, sigma_max = 14.614643096923828
Requested to load SDXL
Loading 1 new model
[Fooocus Model Management] Moving model(s) has taken 1.63 seconds
100%|██████████████████████████████████████████████████████████████████████████████████| 30/30 [00:10<00:00,  2.99it/s]
Requested to load AutoencoderKL
Loading 1 new model
[Fooocus Model Management] Moving model(s) has taken 0.14 seconds
Image generated with private log at: C:\Fooocus_win64_2-1-791\Fooocus\outputs\2024-05-14\log.html
Generating and saving time: 12.61 seconds
[Sampler] refiner_swap_method = joint
[Sampler] sigma_min = 0.0291671771556139, sigma_max = 14.614643096923828
Requested to load SDXL
Loading 1 new model
[Fooocus Model Management] Moving model(s) has taken 1.15 seconds
100%|██████████████████████████████████████████████████████████████████████████████████| 30/30 [00:10<00:00,  2.98it/s]
Requested to load AutoencoderKL
Loading 1 new model
[Fooocus Model Management] Moving model(s) has taken 0.13 seconds
Image generated with private log at: C:\Fooocus_win64_2-1-791\Fooocus\outputs\2024-05-14\log.html
Generating and saving time: 12.14 seconds
Requested to load SDXLClipModel
Requested to load GPT2LMHeadModel
Loading 2 new models
Total time: 25.45 seconds
[Fooocus Model Management] Moving model(s) has taken 0.48 seconds
[Parameters] Adaptive CFG = 7
[Parameters] Sharpness = 2
[Parameters] ControlNet Softness = 0.25
[Parameters] ADM Scale = 1.5 : 0.8 : 0.3
[Parameters] CFG = 4.0
[Parameters] Seed = 8807793496110644830
[Parameters] Sampler = dpmpp_2m_sde_gpu - karras
[Parameters] Steps = 30 - 15
[Fooocus] Initializing ...
[Fooocus] Loading models ...
Refiner unloaded.
[Fooocus] Processing prompts ...
[Fooocus] Encoding positive #1 ...
[Fooocus Model Management] Moving model(s) has taken 0.16 seconds
[Fooocus] Encoding positive #2 ...
[Fooocus] Encoding negative #1 ...
[Fooocus] Encoding negative #2 ...
[Parameters] Denoising Strength = 1.0
[Parameters] Initial Latent shape: Image Space (896, 1152)
Preparation time: 0.73 seconds
[Sampler] refiner_swap_method = joint
[Sampler] sigma_min = 0.0291671771556139, sigma_max = 14.614643096923828
Requested to load SDXL
Loading 1 new model
[Fooocus Model Management] Moving model(s) has taken 1.42 seconds
100%|██████████████████████████████████████████████████████████████████████████████████| 30/30 [00:10<00:00,  2.99it/s]
Requested to load AutoencoderKL
Loading 1 new model
[Fooocus Model Management] Moving model(s) has taken 0.14 seconds
Image generated with private log at: C:\Fooocus_win64_2-1-791\Fooocus\outputs\2024-05-14\log.html
Generating and saving time: 12.45 seconds
[Sampler] refiner_swap_method = joint
[Sampler] sigma_min = 0.0291671771556139, sigma_max = 14.614643096923828
Requested to load SDXL
Loading 1 new model
[Fooocus Model Management] Moving model(s) has taken 1.05 seconds
100%|██████████████████████████████████████████████████████████████████████████████████| 30/30 [00:10<00:00,  2.98it/s]
Requested to load AutoencoderKL
Loading 1 new model
[Fooocus Model Management] Moving model(s) has taken 0.14 seconds
Image generated with private log at: C:\Fooocus_win64_2-1-791\Fooocus\outputs\2024-05-14\log.html
Generating and saving time: 12.02 seconds
Requested to load SDXLClipModel
Requested to load GPT2LMHeadModel
Loading 2 new models
Total time: 25.26 seconds
[Fooocus Model Management] Moving model(s) has taken 0.44 seconds
[Parameters] Adaptive CFG = 7
[Parameters] Sharpness = 2
[Parameters] ControlNet Softness = 0.25
[Parameters] ADM Scale = 1.5 : 0.8 : 0.3
[Parameters] CFG = 4.0
[Parameters] Seed = 9149518165368694657
[Parameters] Sampler = dpmpp_2m_sde_gpu - karras
[Parameters] Steps = 30 - 15
[Fooocus] Initializing ...
[Fooocus] Loading models ...
Refiner unloaded.
[Fooocus] Processing prompts ...
[Fooocus] Encoding positive #1 ...
[Fooocus Model Management] Moving model(s) has taken 0.16 seconds
[Fooocus] Encoding positive #2 ...
[Fooocus] Encoding negative #1 ...
[Fooocus] Encoding negative #2 ...
[Parameters] Denoising Strength = 1.0
[Parameters] Initial Latent shape: Image Space (896, 1152)
Preparation time: 0.51 seconds
[Sampler] refiner_swap_method = joint
[Sampler] sigma_min = 0.0291671771556139, sigma_max = 14.614643096923828
Requested to load SDXL
Loading 1 new model
[Fooocus Model Management] Moving model(s) has taken 1.61 seconds
100%|██████████████████████████████████████████████████████████████████████████████████| 30/30 [00:09<00:00,  3.13it/s]
Requested to load AutoencoderKL
Loading 1 new model
[Fooocus Model Management] Moving model(s) has taken 0.14 seconds
Image generated with private log at: C:\Fooocus_win64_2-1-791\Fooocus\outputs\2024-05-14\log.html
Generating and saving time: 12.11 seconds
[Sampler] refiner_swap_method = joint
[Sampler] sigma_min = 0.0291671771556139, sigma_max = 14.614643096923828
Requested to load SDXL
Loading 1 new model
[Fooocus Model Management] Moving model(s) has taken 1.12 seconds
100%|██████████████████████████████████████████████████████████████████████████████████| 30/30 [00:09<00:00,  3.12it/s]
Requested to load AutoencoderKL
Loading 1 new model
[Fooocus Model Management] Moving model(s) has taken 0.13 seconds
Image generated with private log at: C:\Fooocus_win64_2-1-791\Fooocus\outputs\2024-05-14\log.html
Generating and saving time: 11.39 seconds
Requested to load SDXLClipModel
Requested to load GPT2LMHeadModel
Loading 2 new models
Total time: 24.07 seconds
[Fooocus Model Management] Moving model(s) has taken 0.45 seconds
[Parameters] Adaptive CFG = 7
[Parameters] Sharpness = 2
[Parameters] ControlNet Softness = 0.25
[Parameters] ADM Scale = 1.5 : 0.8 : 0.3
[Parameters] CFG = 4.0
[Parameters] Seed = 4904143005015513720
[Parameters] Sampler = dpmpp_2m_sde_gpu - karras
[Parameters] Steps = 30 - 15
[Fooocus] Initializing ...
[Fooocus] Loading models ...
Refiner unloaded.
[Fooocus] Processing prompts ...
[Fooocus] Encoding positive #1 ...
[Fooocus Model Management] Moving model(s) has taken 0.12 seconds
[Fooocus] Encoding positive #2 ...
[Fooocus] Encoding negative #1 ...
[Fooocus] Encoding negative #2 ...
[Parameters] Denoising Strength = 1.0
[Parameters] Initial Latent shape: Image Space (896, 1152)
Preparation time: 0.49 seconds
[Sampler] refiner_swap_method = joint
[Sampler] sigma_min = 0.0291671771556139, sigma_max = 14.614643096923828
Requested to load SDXL
Loading 1 new model
[Fooocus Model Management] Moving model(s) has taken 1.57 seconds
100%|██████████████████████████████████████████████████████████████████████████████████| 30/30 [00:09<00:00,  3.11it/s]
Requested to load AutoencoderKL
Loading 1 new model
[Fooocus Model Management] Moving model(s) has taken 0.14 seconds
Image generated with private log at: C:\Fooocus_win64_2-1-791\Fooocus\outputs\2024-05-14\log.html
Generating and saving time: 13.19 seconds
[Sampler] refiner_swap_method = joint
[Sampler] sigma_min = 0.0291671771556139, sigma_max = 14.614643096923828
Requested to load SDXL
Loading 1 new model
[Fooocus Model Management] Moving model(s) has taken 1.09 seconds
100%|██████████████████████████████████████████████████████████████████████████████████| 30/30 [00:09<00:00,  3.12it/s]
Requested to load AutoencoderKL
Loading 1 new model
[Fooocus Model Management] Moving model(s) has taken 0.13 seconds
Image generated with private log at: C:\Fooocus_win64_2-1-791\Fooocus\outputs\2024-05-14\log.html
Generating and saving time: 11.82 seconds
Requested to load SDXLClipModel
Requested to load GPT2LMHeadModel
Loading 2 new models
Total time: 25.56 seconds
[Fooocus Model Management] Moving model(s) has taken 0.49 seconds

Additional information

Using juggernautXL_v8Rundiffusion.safetensors

[mashb1t]

Hey @llagerlof, the console log you've attached is unrelated and contains a prompt for "blue bird standing still". Please provide the metadata of the image (history log copy button) so this can be debugged 1:1 as you've created the image. Thanks!

[llagerlof]

@mashb1t, I've been able to reproduce today:

{
"prompt": "dog walking, german shepherd breed",
"negative_prompt": "",
"prompt_expansion": "",
"styles": "['Flat 2d Art']",
"performance": "Speed",
"resolution": "(1152, 896)",
"guidance_scale": 4,
"sharpness": 2,
"adm_guidance": "(1.5, 0.8, 0.3)",
"base_model": "juggernautXL_v8Rundiffusion.safetensors",
"refiner_model": "None",
"refiner_switch": 0.5,
"sampler": "dpmpp_2m_sde_gpu",
"scheduler": "karras",
"seed": "6582132752346538731",
"lora_combined_1": "sd_xl_offset_example-lora_1.0.safetensors : 0.1",
"metadata_scheme": false,
"version": "Fooocus v2.3.1"
}

About the previous console log I posted, I don't know why it's appearing an old prompt. You can ignore it. You probably can reproduce the problem with this prompt configuration.

[mashb1t]

Please check if adding these keywords helps:

positive: sharp edges, minimalistic
negative: depth, high details

This worked for me, but some seeds in general work better than others with specific models.
In general, I'd propose not to touch the style as users already may have a collection of images which they will then not be able to reproduce again, but rather add a new style.
You may also try to use the style "Papercraft Flat Papercut", but this may be too extreme and colorless, at least without tweaking it a bit.

What do you think?

[xhoxye]

Set this parameter to 7，guidance_scale CFG = 7，Because "juggernautXL_v8" CFG=4 is for photos

[mashb1t]

@llagerlof is this issue resolved and were you able to modify the prompt to your liking?

[llagerlof]

@llagerlof is this issue resolved and were you able to modify the prompt to your liking?

Well, not always, but I think you can close this issue. Probably this is a thing to be fixed on model training level, I don't know.

Thank you.