[a / b / c / d / e / f / g / gif / h / hr / k / m / o / p / r / s / t / u / v / vg / vm / vmg / vr / vrpg / vst / w / wg] [i / ic] [r9k / s4s / vip / qa] [cm / hm / lgbt / y] [3 / aco / adv / an / bant / biz / cgl / ck / co / diy / fa / fit / gd / hc / his / int / jp / lit / mlp / mu / n / news / out / po / pol / pw / qst / sci / soc / sp / tg / toy / trv / tv / vp / vt / wsg / wsr / x / xs] [Settings] [Search] [Mobile] [Home]
Board
Settings Mobile Home
/g/ - Technology


Thread archived.
You cannot reply anymore.


[Advertise on 4chan]


File: tmp.jpg (1.09 MB, 3264x3264)
1.09 MB
1.09 MB JPG
Discussion of free and open source text-to-image models

Previous /ldg/ bread : >>101720035

>Beginner UI
EasyDiffusion: https://easydiffusion.github.io
Fooocus: https://github.com/lllyasviel/fooocus
Metastable: https://metastable.studio

>Advanced UI
Automatic1111: https://github.com/automatic1111/stable-diffusion-webui
ComfyUI: https://github.com/comfyanonymous/ComfyUI
InvokeAI: https://github.com/invoke-ai/InvokeAI
SD.Next: https://github.com/vladmandic/automatic
SwarmUI: https://github.com/mcmonkeyprojects/SwarmUI

>Use a VAE if your images look washed out
https://rentry.org/sdvae

>Model Ranking
https://imgsys.org/rankings

>Models, LoRAs & training
https://civitai.com
https://huggingface.co
https://aitracker.art
https://github.com/Nerogar/OneTrainer
https://github.com/derrian-distro/LoRA_Easy_Training_Scripts

>Flux
https://huggingface.co/spaces/black-forest-labs/FLUX.1-schnell
https://comfyanonymous.github.io/ComfyUI_examples/flux

>Pixart Sigma & Hunyuan DIT
https://huggingface.co/spaces/PixArt-alpha/PixArt-Sigma
https://huggingface.co/spaces/Tencent-Hunyuan/HunyuanDiT
https://huggingface.co/comfyanonymous/hunyuan_dit_comfyui
Nodes: https://github.com/city96/ComfyUI_ExtraModels

>Kolors
https://gokaygokay-kolors.hf.space
Nodes: https://github.com/kijai/ComfyUI-KwaiKolorsWrapper

>AuraFlow
https://fal.ai/models/fal-ai/aura-flow
https://huggingface.co/fal/AuraFlows

>Index of guides and other tools
https://rentry.org/sdg-link
https://rentry.org/rentrysd

>GPU performance
https://vladmandic.github.io/sd-extension-system-info/pages/benchmark.html
https://docs.getgrist.com/3mjouqRSdkBY/sdperformance

>Try online without registration
txt2img: https://www.mage.space
img2img: https://huggingface.co/spaces/huggingface/diffuse-the-rest
sd3: https://huggingface.co/spaces/stabilityai/stable-diffusion-3-medium

>Related boards
>>>/h/hdg
>>>/e/edg
>>>/d/ddg
>>>/b/degen
>>>/vt/vtai
>>>/aco/sdg
>>>/trash/sdg
>>
Blessed thread of frenship
>>
File: based5.jpg (267 KB, 715x900)
267 KB
267 KB JPG
>Blessed thread of frenship
>>
File: ComfyUI_30724_.png (1.55 MB, 1024x1024)
1.55 MB
1.55 MB PNG
>>
>>
File: ComfyUI_00097_.png (851 KB, 1152x896)
851 KB
851 KB PNG
>>
File: ComfyUI_00098_.png (1.16 MB, 896x1152)
1.16 MB
1.16 MB PNG
>>
So in a couple days we went from "Flux is impossible to make loras for and it will NEVER happen" to "it may be possible but you need 75gb of VRAM" to "okay just 35 but that's still nonviable" to 24 to 18 and now targeting 16 and maybe even less. It's just a matter of time until it's fine-tunable with a 3090. SD is over.
>>
File: basedboy.jpg (456 KB, 1296x1654)
456 KB
456 KB JPG
>>Blessed thread of frenship
>>
File: 1707764124758649.png (1.26 MB, 1024x1024)
1.26 MB
1.26 MB PNG
finally, the miku series I have been waiting for.
>>
File: file.png (185 KB, 556x600)
185 KB
185 KB PNG
Guys i'm scared... Momo drew this
>>
File: ComfyUI_00099_.png (1.02 MB, 896x1152)
1.02 MB
1.02 MB PNG
>>
>>101725162
Okay but how is the AMD support
>>
File: FLUX__00164_.png (1.41 MB, 1024x1024)
1.41 MB
1.41 MB PNG
>>
>>101725216
She's asking for it.
>>
>>101725162
We still need to see a final product of one of those loras. There were a couple of cope settings to train a lora on XL with 8GB when it released, but there was clearly a quality cost attached.
>>
File: FLUX__00165_.png (1.38 MB, 1024x1024)
1.38 MB
1.38 MB PNG
>>
File: FD_00088_.png (1.79 MB, 1024x1024)
1.79 MB
1.79 MB PNG
What if daredevil was superman
>>
>>101725236
My experience has been that 12 gb is the bare minimum to train a good style lora for pony, so I frankly doubt that will be enough to train for flux
>>
File: 1691402853025080.png (1.2 MB, 1024x1024)
1.2 MB
1.2 MB PNG
World of Warcraft ingame screenshot, UI and HUD visible, the main character is Donald Trump, he is wearing plate armor and is holding a large greatsword. The background is Icecrown Citadel from World of Warcraft. The action bar at the bottom has sword icons. The minimap at the top right is a map of Icecrown.
>>
File: fr248.jpg (155 KB, 1024x1024)
155 KB
155 KB JPG
>>101725248
he is
>>
File: 1694291597966333.png (1.2 MB, 1024x1024)
1.2 MB
1.2 MB PNG
>>101725264
>>
File: ComfyUI_00101_.png (1.07 MB, 896x1152)
1.07 MB
1.07 MB PNG
>>
File: FD_00241_.png (1.74 MB, 1024x1024)
1.74 MB
1.74 MB PNG
>>101725268
>French fry burrito
You Americans will eat anything.
Also Flux is good at food
>>
>>101725216
that's pretty good.. what's flux?
>>
File: ComfyUI_01987_.png (1.78 MB, 1280x1024)
1.78 MB
1.78 MB PNG
>>
File: ComfyUI_30726_.png (1.37 MB, 1024x1024)
1.37 MB
1.37 MB PNG
>>
File: ComfyUI_01069_.png (1.1 MB, 1024x1024)
1.1 MB
1.1 MB PNG
>>
File: sok2.jpg (107 KB, 1024x1024)
107 KB
107 KB JPG
>>101725296
i'm as American as you
>>
File: ComfyUI_01973_.png (1.77 MB, 1280x1024)
1.77 MB
1.77 MB PNG
>>
File: ComfyUI_01070_.png (934 KB, 1024x1024)
934 KB
934 KB PNG
>>101725331
>>
>>101725331
hey that's my mom
>>
>>101725317
>>101725341
Love the naga
>>
File: ComfyUI_01071_.png (1.26 MB, 1024x1024)
1.26 MB
1.26 MB PNG
>>
File: 1703546826462542.png (1.28 MB, 1024x1024)
1.28 MB
1.28 MB PNG
>>101725286
Miku edition:
>>
File: ComfyUI_30727_.png (1.55 MB, 1024x1024)
1.55 MB
1.55 MB PNG
>>
Upscale from 1024x1024 to 2048x2048 rather than directly genning at 2048 is fine. But fuck me is it slow.
>>
File: FLUX__00084_.png (1.17 MB, 1024x1024)
1.17 MB
1.17 MB PNG
>>
>>101725396
We can upscale with flux huh? Can it inpaint too?
>>
>>101725396
Upscale using?
>>
>>101725396
Forgot pic
>>101725403
Haven't tried inpainting mostly because I hate inpainting in comfy.
>>101725409
4x ultra sharp
>>
File: ComfyUI_30728_.png (1.43 MB, 1024x1024)
1.43 MB
1.43 MB PNG
>>
File: ComfyUI_01072_.png (1.11 MB, 1024x1024)
1.11 MB
1.11 MB PNG
>>
File: FD_00151_.jpg (921 KB, 2048x2048)
921 KB
921 KB JPG
>>101725422
>>101725396
>Forgot pic
I am a retard
>>
>>101725433
did you denoise it after upscaling? if so by how much? how much more vram on the upscale?
>>
File: image (1).jpg (204 KB, 1024x1024)
204 KB
204 KB JPG
Swimsuit made out of guts
>>
>>101725433
Yeah ran another ksampler at 0.2 noise with the same seed
>>
>>101725030
Been running into all sorts of issues while trying to get sd working locally on my machine (AMD/Linux). The guide on the stable-diffusion-webui-amdgpu repo linked to in the rentry up above has gotten me the closest so far, and while the webui does now install/launch without errors, I can't seem to get it to actually start generating anything. The buttons in the webui seem to not work and there's no new output in the console after start up. Any ideas what might be causing this?
inb4 no model, this guide installed v1-5-pruned-emaonly.safetensors for me, but I also tried two other models from civitai without any luck.
>>
File: image (21).jpg (221 KB, 1024x1024)
221 KB
221 KB JPG
>>
>>101725473
see >>101725516
I am fucking brain damaged today I don't know what is wrong with me
>>
>>101725531
You could try comfy if auto is flat out not working. I don't use AMD/Linux so I have no clue what the issue might be. You should ask /sdg/ too in case someone there does.
>>
>>101725531
Are the other models you are trying also 1.5 models or are they something else?
>>
>>101725399
smother me in them titties
>>
File: ComfyUI_Flux_2955.jpg (195 KB, 1024x768)
195 KB
195 KB JPG
>>
>>101725631
One of them is based on SDXL 1.0, but the other was based on 1.5.
>>
>>101725531
You have to get amd working right first. Then you have to get Python working right with Pytorch.

In Ubuntu, you have to use a venv, or at least I do.
>>
File: ComfyUI_temp_zorkh_00002_.jpg (2.67 MB, 2056x2056)
2.67 MB
2.67 MB JPG
I'm doing a 2-step high res images, but the second sampler adds this shitty looking noise, what is fucking up? Both samplers are euler, normal, 4 steps.
>>
File: 1718027131784279.png (1.21 MB, 1024x1024)
1.21 MB
1.21 MB PNG
>>
File: ComfyUI_30735_.png (1.13 MB, 1024x1024)
1.13 MB
1.13 MB PNG
>>
>>101725701
The guide I followed created a venv and I am makings sure to activate it prior to launching webui.
>>
File: 1717392003493284.png (1.21 MB, 1024x1024)
1.21 MB
1.21 MB PNG
what a time to be alive, and they said dall-e would never be open source.
>>
>>101725699
get the 1.5 working first, I did.
>>
>>101725714
did you install everything with pip from install.txt like it said?
>>
what guidance do you guys use for flux? I'm still not clear on what changes if you raise or lower from the default 3.5
>>
File: ComfyUI_Flux_2971.jpg (205 KB, 1024x768)
205 KB
205 KB JPG
>>
>>101725702
Was your guidance extra low when you created the first image?
>>
File: 1702656584997523.png (1.23 MB, 1024x1024)
1.23 MB
1.23 MB PNG
>>101725719
pixar style:
>>
File: 1722810749.png (1.72 MB, 1024x1024)
1.72 MB
1.72 MB PNG
>>
>>101725739
It was at 1 yes
>>
File: 1715320693402375.png (1.26 MB, 1024x1024)
1.26 MB
1.26 MB PNG
>>101725743
>>
File: ComfyUI_Flux_2973.jpg (176 KB, 1024x768)
176 KB
176 KB JPG
>>101725734
2-2.5
>>
>>101725727
Is the v1-5-pruned-emaonly that came with my install not 1.5? Regardless, I'll give manually downloading it a try.

>>101725731
The install script (I believe) should handle that for me, but regardless, I already tried manually installing all the packages using the install.txt, but I just get the same result.
>>
File: ComfyUI_00008_.png (704 KB, 1024x768)
704 KB
704 KB PNG
>>
File: FD_00164_.png (1.67 MB, 1024x1024)
1.67 MB
1.67 MB PNG
We modeled our armor to cover the areas where soldiers returning from battle are most commonly wounded.
>>
File: ComfyUI_30737_.png (1.28 MB, 1024x1024)
1.28 MB
1.28 MB PNG
>>
File: ComfyUI_Flux_2979.jpg (237 KB, 1024x768)
237 KB
237 KB JPG
>>
File: flux celebs.jpg (2.42 MB, 5760x2560)
2.42 MB
2.42 MB JPG
hmm
>>
>>101725765
neat
>>
File: 1701983813728252.png (1.12 MB, 1024x1024)
1.12 MB
1.12 MB PNG
>>
File: Flux_00275_.png (1.07 MB, 1152x768)
1.07 MB
1.07 MB PNG
>>
File: ComfyUI_Flux_2983.jpg (179 KB, 1024x768)
179 KB
179 KB JPG
>>
File: FD_00500_.png (1.22 MB, 1024x1024)
1.22 MB
1.22 MB PNG
>>101725781
How do you think they made it safe, Anon?
Also, they are definitely still in there. Only their names were scrubbed.
>>
File: FD_00166_.png (1.45 MB, 1024x1024)
1.45 MB
1.45 MB PNG
>>101725781
>>101725800
>>
>>101725781
>suit, vans/converse shoes
>>
File: FD_00167_.png (1.24 MB, 1024x1024)
1.24 MB
1.24 MB PNG
>>101725822
>>
File: 1704979070239770.png (1.26 MB, 1024x1024)
1.26 MB
1.26 MB PNG
open source wins cause dall-e would reject this prompt.
>>
File: ComfyUI_30739_.png (1.32 MB, 1024x1024)
1.32 MB
1.32 MB PNG
>>
>>101725762
I downloaded exactly the v1-5 whatever and put it under ComfyUI/models/checkpoints
>>
>>101725800
names scrubbed how? you can prompt for swift
>>
File: 1722811395.png (1.49 MB, 1024x1024)
1.49 MB
1.49 MB PNG
>>
>>101725860
>>101725762
>>
>>101725872
Not everything was scrubbed.
Just like you can get nudes yet they scrubbed most of them.
>>
>>101725877
Can you reverse it? A giant cockroach with a plate of humans?
>>
File: ComfyUI_30740_.png (1.12 MB, 1024x1024)
1.12 MB
1.12 MB PNG
>>
>>101725872
You can probably gen taytay because her name is ambiguous, not clearly female. I have a feeling their VLM just scrubbed female names, rather than all celebrity names, so it's harder for us to gen them fucking donkeys and harder for BFL to get sued.
Nobody cares if we gen a horse cock onto George Clooney.
>>
File: Flux_00293_.png (979 KB, 1152x768)
979 KB
979 KB PNG
it knows what trollface is but they pruned pepe the frog from the dataset. sadness.
>>
>>101725895
do you prompt pixel art at the start?
>>
>>101725895
Neat, prooompt?
>>
>>101725531
I was having the exact same issues (on debian), switching to comfyui solved most of it and from there you can just steal someone's workflow and build off it yourself.
Had to install pyenv and python 3.10.6 through it first, then set
pyenv shell 3.10.6
before installing anything else related to comfyui. automatic1111 wouldn't even recognize my gpu no matter what I tried.
>>
File: ComfyUI_01078_.png (900 KB, 1024x1024)
900 KB
900 KB PNG
>>
File: ComfyUI_01079_.png (982 KB, 1024x1024)
982 KB
982 KB PNG
>>
File: ComfyUI_30741_.png (1.46 MB, 1024x1024)
1.46 MB
1.46 MB PNG
>>101725916
>>101725933
https://files.catbox.moe/kcyt48.png
The prompt is too long to post. I accidentally pasted the prompt I gave the text gen AI as the first line.
>>
>>101725888
https://files.catbox.moe/kud2ta.png

Looks more like barbie dolls and cockroach wasn't that huge, so it sort of works. Maybe I need to try it in a different context.
>>
>>101725995
Pretty funny though, maybe realistic cockroach man?
>>
>>101725755
how do I get stuff like this?
>>
File: ComfyUI_00144_.png (824 KB, 1344x768)
824 KB
824 KB PNG
>>
File: 1722811804.png (1.24 MB, 1024x1024)
1.24 MB
1.24 MB PNG
>>
File: ComfyUI_00271_ (1).png (1.18 MB, 1024x1024)
1.18 MB
1.18 MB PNG
>>
File: ComfyUI_Flux_3015.jpg (173 KB, 1024x768)
173 KB
173 KB JPG
>>101725888
nta and not really humans
>>
>>101726008
give it a style like pixar and describe the text and location

prompt is like:

The title "POO in the LOO" is visible in bold and playful text. The background is Mumbai, India. Several Indian men wearing a Turban are squatting in the street. Poop is scattered around the road. The poop looks cartoonish. The image style is like a Pixar film. At the bottom the text "just loo it" is visible. Include Pixar branding to make it look like a Pixar movie poster. The Pixar logo is visible at the bottom of the image.
>>
File: ComfyUI_00116_.png (1.07 MB, 1024x1024)
1.07 MB
1.07 MB PNG
I love flux
>>
File: ComfyUI_temp_zorkh_00006_.jpg (1.62 MB, 1816x1816)
1.62 MB
1.62 MB JPG
>>
File: ComfyUI_30743_.png (1.5 MB, 1024x1024)
1.5 MB
1.5 MB PNG
>>
File: ComfyUI_01080_.png (994 KB, 1024x1024)
994 KB
994 KB PNG
>>
>>101725531
I haven't tried webui but I have no issues with comfy. What AMD card are you using?
>>
File: 1714171776139890.png (1.4 MB, 1024x1024)
1.4 MB
1.4 MB PNG
the custom children's style fonts are making me laugh, lmao

The title "Dora the Auschwitz Explorer" is visible in bold and playful text. The background is Auschwitz. Dora the Explorer is standing in front of Auschwitz with a smile on her face. She is running towards the entrance. A sign saying "Arbeit Macht Frei" is visible on a steel gate. The image style is like a Pixar film. The Disney logo is at the top of the image. Include a tagline that reads: "let's explore!" prominently on the poster. The visual style is in the style of a Disney movie.

if it can do edgy prompts, anything is possible.
>>
File: Flux_00311_.png (1.02 MB, 1152x768)
1.02 MB
1.02 MB PNG
>>
File: ComfyUI_30744_.png (1.38 MB, 1024x1024)
1.38 MB
1.38 MB PNG
>>
File: ComfyUI_Flux_3033.jpg (197 KB, 1024x768)
197 KB
197 KB JPG
>>101725888
>>101726025
improved it
>>
File: ComfyUI_00259_.png (844 KB, 1024x1024)
844 KB
844 KB PNG
>>
>>101726074
do you prompt pixel art at the start to get that style?
>>
>>101726084
Amazing.
>>
What the fuck will this look like in another couple of years
>>
>>101726098
b& for your own safety. You'll be able to rent AI time from trusted companies.
>>
File: ComfyUI_01081_.png (1.01 MB, 1024x1024)
1.01 MB
1.01 MB PNG
>>
File: ComfyUI_00012_.png (1.2 MB, 768x1280)
1.2 MB
1.2 MB PNG
>>
File: ComfyUI_00270_.png (918 KB, 1024x1024)
918 KB
918 KB PNG
>>
File: file.png (739 KB, 676x666)
739 KB
739 KB PNG
>>101726030
for some reason AI doesn't like the word pedophile
>>
File: ComfyUI_Flux_3035.jpg (191 KB, 1024x768)
191 KB
191 KB JPG
coming home after a hard day of work to a plate of steaming humans
>>
>>101726124
lmao top kek
>>
File: ComfyUI_Flux_3039.jpg (221 KB, 1024x768)
221 KB
221 KB JPG
>>
>>101726096
yeah i just began the prompt with pixel art
it's a gamble as to what magnification you will get, since these models are not trained to be pixel perfect
>>
File: ComfyUI_16166_.png (1.87 MB, 768x1344)
1.87 MB
1.87 MB PNG
>>
>>101726140
Nice, the size is right!
>>
File: ComfyUI_01084_.png (1.17 MB, 1024x1024)
1.17 MB
1.17 MB PNG
>>
File: ComfyUI_00118_.png (1.06 MB, 1024x1024)
1.06 MB
1.06 MB PNG
sucks when it's almost right and you gotta wait a while to gen another try
>>
File: 1717322454951924.png (1.16 MB, 1024x1024)
1.16 MB
1.16 MB PNG
I might cause an international incident if this is posted
>>
>>101726166
lmao it will do anything.
>>
File: 1710649294165732.png (1.15 MB, 1024x1024)
1.15 MB
1.15 MB PNG
>>101726162
>>
File: 1722813076.png (1.72 MB, 1024x1024)
1.72 MB
1.72 MB PNG
>>
>>101726162
what GPU u using m8? how long your gens take?
what do you do while it generates?
PC is basically unusable for me when it generates.
>>
File: Flux_00312_.png (904 KB, 1152x768)
904 KB
904 KB PNG
>>
File: ComfyUI_temp_zorkh_00012_.png (3.86 MB, 1816x1816)
3.86 MB
3.86 MB PNG
>>
File: ComfyUI_01086_.png (1.08 MB, 1024x1024)
1.08 MB
1.08 MB PNG
>>
File: 1699847934818546.png (109 KB, 1417x811)
109 KB
109 KB PNG
>>101726173
yep, and this is why open source always wins. even the new llama 3.1 LLM models have uncensored versions now.

what a time to be alive, huh?
>>
File: FD_00204_.png (983 KB, 1024x1024)
983 KB
983 KB PNG
>>101726162
>>
>>101726063
>if it can do edgy prompts, anything is possible.
It can't do nsfw.
>>
>>101726189
rtx 3060, 2-3 mins for a 1024x1024 gen, PC is just fine while genning. If your PC is freezing up I suspect you are running low on RAM, which is required by the large CLIP models for flux.
>>
>>101726227
it can't do full frontal or porn. It does erotica and gore just fine
>>
File: ComfyUI_01087_.png (1.31 MB, 1024x1024)
1.31 MB
1.31 MB PNG
>>
>>101726227
Neither could any of the other base models beyond ass and titties.
>>
>>101726236
I use a 4090 and have 64GB of RAM, PC isnt actually freezing up but browsing with 100 tabs in chrome is somewhat choppy and cant really play any videogames or watch 4 videos on jewtube while waiting.
its a pain in the ass.
>>
File: Flux_00320_.png (964 KB, 1152x768)
964 KB
964 KB PNG
>>101726227
i don't see the appeal of nsfw that much, the images don't turn me on because i know they're not real
>>
File: ComfyUI_01088_.png (1.29 MB, 1024x1024)
1.29 MB
1.29 MB PNG
>>101726274
I think its alright that it cant into nsfw for now, it forces people to be more creative.
>>
>>101726227
if you really want tits, just take a flux gen and inpaint/img2img with sdxl or ponyxl models.
>>
File: ComfyUI_temp_uiyoi_00041_.png (1.54 MB, 1400x1400)
1.54 MB
1.54 MB PNG
>>
>>101726271
that's when you bust out the laptop for your second screen
>>
File: ComfyUI_00119_.png (2.25 MB, 1536x1536)
2.25 MB
2.25 MB PNG
>>
File: ComfyUI_temp_uiyoi_00042_.png (2.71 MB, 1400x1400)
2.71 MB
2.71 MB PNG
>>
File: ComfyUI_30747_.png (820 KB, 1024x1024)
820 KB
820 KB PNG
I remember struggling with bats on SD and eventually putting them into negatives, flux gens them very nice even as pixel art.
>>
File: 1697833178200928.png (63 KB, 637x757)
63 KB
63 KB PNG
>>101726236
>>101726271

these are my settings and I get good gens in 25-30 seconds with a 4080 and 32gb ram. fp16 weight kills efficiency though and you definitely need a 4090 or more than 16GB VRAM.
>>
>>101726293
for sure, I gotta upgrade my second PC.
>>
File: 1721273367651354.png (1.05 MB, 1024x1024)
1.05 MB
1.05 MB PNG
open source always wins.
>noooooooo you cant prompt that!
>please pay for tokens to generate images!
>>
>>101726311
even 24GB vram isnt enough nowadays.
>>
>>101726311
that's pretty good, i measured 24 seconds per image on my 3090 with fp8
>>
File: ComfyUI_01090_.png (1012 KB, 1024x1024)
1012 KB
1012 KB PNG
>>
File: ComfyUI_temp_uiyoi_00045_.png (2.56 MB, 1400x1400)
2.56 MB
2.56 MB PNG
>>
File: FLUX__00004_.png (974 KB, 896x1152)
974 KB
974 KB PNG
can't get it to do water, wetness, moistness, humidness, or anything like that
no ESL bully please
>>
>>101726311
what about all the other comfy UI settings like steps, samplers and schedulers and shit?
any good combos?
>>
File: ComfyUI_00120_.png (1.36 MB, 1024x1024)
1.36 MB
1.36 MB PNG
>>101726311
I don't get it, everyone keeps saying you need lots of VRAM but I never have issues genning full fp16 flux1-dev on my 3060. Kinda slow at 2 or 3 mins per gen but I can hardly complain considering the quality.
>>
I fucking love flux. I don't even know if finetunes are absolutely necessary as long as we get LoRAs and Controlnets.
>>
>>101726311
i found changing the dtype from default to fp8 ran like absolute shit on my 3080. from about 1min to 10-15min. ram usage went way down though.
>>
File: ComfyUI_temp_uiyoi_00046_.png (2.3 MB, 1400x1400)
2.3 MB
2.3 MB PNG
KINO
>>
>>101726384
I didn't mess with anything else. 20 steps for dev model. If you use the schnell model, you can get results in 1-4 steps even, think of it like SDXL turbo.
>>
>>101726393
Are you seeing
>Loading in lowvram mode
popping up?
>>
>>101726394
There are some loras on HF. So I think its doable.

I'm just wondering if we can get a quantized versions. Q4 Q5.
>>
File: ComfyUI_01093_.png (977 KB, 1024x1024)
977 KB
977 KB PNG
>pulls you over
>"booboo gaga gugu"
>>
>>101726403
Schnell only saves me about 10~20seconds and the quality (especially in text) is far worse for me. Unless I'm doing it wrong and need to use a different sampler or something.
>>
>>101726425
I think the dev model but with fp8 weight is the best mix of speed/quality so far, imo: but I don't have 24gb VRAM.
>>
File: Flux_00076_.png (985 KB, 1104x904)
985 KB
985 KB PNG
>>101726162
>>
>>101726406
>Loading in lowvram mode
I see this and I run a 4090, apparently 24GB is "lowvram"
>>
>>101726451
We all get to be poorfags together.
>>
File: flux.jpg (136 KB, 1024x1024)
136 KB
136 KB JPG
>>101726227
It can, it's definitely in there.
>>
>>101726443
I'll be perfectly honest. The FP8 outputs aren't that much worse than the FP16 ones. It's like a line here or there difference.
>>
File: ComfyUI_Flux_3103.jpg (233 KB, 1024x768)
233 KB
233 KB JPG
>>
File: ComfyUI_00121_.png (1.41 MB, 1024x1024)
1.41 MB
1.41 MB PNG
>>101726406
Yeah, but that comes up for fp8 as well and the speed is basically the same as fp16. And terminal output confirms it is running in bfloat16
>>
File: ComfyUI_00275_ (1).png (1.15 MB, 1024x1024)
1.15 MB
1.15 MB PNG
>>101726173
yes, anything indeed
>>
>>101726303
did yu prompt thes hirt
>>
File: 1722296305361027.png (1.34 MB, 1024x1024)
1.34 MB
1.34 MB PNG
this is what DALL-E doesnt want you to know:
>>
>>101726484
ye, just put "wearing a pink floyd shirt"
>>
>>101726472
Just remnants of it, at most artistic nudity or missed images from the training data set. And besides, nipples are fucked, there is almost no nude genitals, almost no erotica (as in lingerie etc outside of tame magazines).
>>
File: cuda.png (24 KB, 679x571)
24 KB
24 KB PNG
>>101726406
Not him, but I do (3060), also my CUDA graph looks like this
>>
File: ComfyUI_Flux_3109.jpg (202 KB, 1024x768)
202 KB
202 KB JPG
>>
>>101726425
>>101726443
Same experience here on my 2060 12GB. 460 seconds on Schnell vs 470 seconds on dev-fp8, much more accurate on the latter.
>>
File: ComfyUI_30748_.png (1.14 MB, 1024x1024)
1.14 MB
1.14 MB PNG
>>
>>101726530
like castlevania on the NES, cool
>>
>>101726471
really shows how much technology is lagging behind for all this AI shit.
>>
File: ComfyUI_00122_.png (1.97 MB, 1024x1024)
1.97 MB
1.97 MB PNG
>>
File: FD_00221_.png (1.43 MB, 1024x1024)
1.43 MB
1.43 MB PNG
I love that you can just prompt for action and poses with Flux, no control net. No more 1girl staring vacantly into the distance
>>101726511
It's fixed with fine tuning. Base 1.5 and base XL suck at nudes too.
You can wait for the finetunes or you can continue to bitch like a little faggot.
>>
File: 1691768501822374.png (1.24 MB, 1024x1024)
1.24 MB
1.24 MB PNG
you are literally only limited by your imagination.
>>
>>101726556
needs a punchier title, the line itself can be the tagline
>>
File: FLUX__00009_.png (1 MB, 896x1152)
1 MB
1 MB PNG
>>
>>101726556
>limited by your imagination
ironic
>>
>>101726555
>It's fixed with fine tuning. Base 1.5 and base XL suck at nudes too.
We'll see if people even can pay for that.

>You can wait for the finetunes or you can continue to bitch like a little faggot.
I'm just sad that it could have been amazing from the get go anon.
I tried asking for characters to have fun poses, it almost recognizes no one, artists it's the same, even public domain ones.
>>
File: 0086.jpg (70 KB, 920x648)
70 KB
70 KB JPG
>>
>>101726591
If you can't prompt without naming artists you are a promptlet, simple as that.
>>
>>101726556
>Pixar style
>Edgy Alex Jones or jews or nazis or nigger text
Yeah.
>>
File: 1702119912600295.png (1.16 MB, 1024x1024)
1.16 MB
1.16 MB PNG
>>101726590
if it can generate stuff that dall-e rejects cause of "ethics" then you can literally do whatever you like, even generate isometric videogame graphics.
>>
>>101726598
Not what I wrote.
>>
File: 1715875562307637.png (1.26 MB, 1024x1024)
1.26 MB
1.26 MB PNG
im impressed by the reflections desu
>>
File: ComfyUI_30738_.png (1.22 MB, 1024x1024)
1.22 MB
1.22 MB PNG
>>101726556
Imagination is overrated, just ask mistral nemo something like
>very detailed description of a very insensitive, offensive, crude and politically incorrect videogame cover, specify the title, the slogan and their placement
>>
File: ComfyUI_01095_.png (1.45 MB, 1024x1024)
1.45 MB
1.45 MB PNG
>>
>>101726630
Very nice
>>
File: FD_00240_.png (1.34 MB, 1024x1024)
1.34 MB
1.34 MB PNG
>>101726591
I dunno what to tell you man. It is a BASE model.
We have this same argument literally every time a BASE model comes out. And every time, every single complaint is fixed with tuning. At what point do you give up and realise your complaint is both a non-issue and stupid?
>>
File: ComfyUI_01096_.png (1.45 MB, 1024x1024)
1.45 MB
1.45 MB PNG
>>101726639
>>
>>101726663
This gives a planet of the ape timeline vibe, good stuff.
>>
File: ComfyUI_30750_.png (1.21 MB, 1024x1024)
1.21 MB
1.21 MB PNG
>>
File: ComfyUI_30751_.png (883 KB, 1024x1024)
883 KB
883 KB PNG
>>
File: file.png (375 KB, 551x534)
375 KB
375 KB PNG
what a lucky man
>>
>>101726687
this is where white people came from
>>
>>101726694
I believe it
>>
File: 1718461083005825.png (1.1 MB, 1024x1024)
1.1 MB
1.1 MB PNG
>in game screenshot of fallout 3, the main character is Miku Hatsune
>>
>>101726675
This looks fantastic, the fact that there's coherence between those little spike cluster sprites is amazing. Catbox?
>>
File: ComfyUI_30752_.png (1.08 MB, 1024x1024)
1.08 MB
1.08 MB PNG
>>
File: 1722815658602188.jpg (111 KB, 1024x1024)
111 KB
111 KB JPG
It's him, the 'ecker
>>
File: 1693007020408465.png (1.49 MB, 1024x1024)
1.49 MB
1.49 MB PNG
pixel art. Final Fantasy 6 in game screenshot. UI and HUD visible. The main character is Hatsune Miku. The setting is a medieval castle.
>>
Will you be joining the competition?
>>
File: ComfyUI_01100_.png (1.51 MB, 1024x1024)
1.51 MB
1.51 MB PNG
>>101726668
>>
>>101726714
>>101726707
Can you generate something that isn't symmetrical
>>
File: ComfyUI_00235_.png (1.23 MB, 1024x1024)
1.23 MB
1.23 MB PNG
>>
File: 1697419439393087.png (987 KB, 1024x1024)
987 KB
987 KB PNG
pixel art. Chrono Trigger in game screenshot. UI and HUD visible. The main character is Hatsune Miku. The setting is a forest.
>>
>>101726514
bottleneck because leaking to ram
>>
File: 1711548170284704.png (1.62 MB, 1024x1024)
1.62 MB
1.62 MB PNG
>>
File: ComfyUI_00245_ (1).png (1.09 MB, 1024x1024)
1.09 MB
1.09 MB PNG
>>
File: ComfyUI_30753_.png (891 KB, 1024x1024)
891 KB
891 KB PNG
>>101726704
https://files.catbox.moe/6hv0a0.png
>>101726740
It all depends on whatever wall of text I get from mistral nemo.
>>
File: ComfyUI_01102_.png (1.38 MB, 1024x1024)
1.38 MB
1.38 MB PNG
>>
>>101726209
Nice, which version is this?
>>
>>101726785
llama 3.1 uncensored, download LM studio (app to load the models, it's small) and then find it in the file browser, or google llama 3.1 uncensored models.
>>
>>101726773
Thank you bro
>>
>>101726778
hogs of war reboot?
>>
File: ComfyUI_01104_.png (1.53 MB, 1024x1024)
1.53 MB
1.53 MB PNG
>>101726804
>>
File: peaceman.png (1.34 MB, 1024x1024)
1.34 MB
1.34 MB PNG
>>101726483
>>
>>101726796
*Also you need 50gb of vram to run the good models at a reasonable speed* teehee
>>
File: 1720534800093592.png (1.42 MB, 1024x1024)
1.42 MB
1.42 MB PNG
>>
File: ComfyUI_00129_.png (803 KB, 1024x1024)
803 KB
803 KB PNG
>>
>>101726822
nah you can get the 8b model that is like 4 gigs and it works great, dont get the giga hueg ones.
>>
File: ComfyUI_00397_.png (742 KB, 1344x768)
742 KB
742 KB PNG
>>
File: ComfyUI_01105_.png (1.28 MB, 1024x1024)
1.28 MB
1.28 MB PNG
>>
>>101726832
Nah the 8b ones just don't get it. Sure they can do basic tasks but they can't understand the situation in the same way the larger models do.
>>
>>101726548
Cool, what proompt?
>>
>>101726282
creative, exactly my humor.
there's a reason why 95% of the SD ecosystem consists of porn.
now people are generating thousands and thousands of memes for 2 weeks that nobody cares about because everyone thinks their meme is better. *slap
no future without porn
>>
>>101726556
Did you like The School That Never Was? Well, Alex Jones is back with his sequel, Frogwater.
>>
File: ComfyUI_00022_.png (838 KB, 1024x1024)
838 KB
838 KB PNG
I'm trying to generate a Ted Kazynski funko pop, but every time it spells Kaczynski correctly the image is incredibly low quality. I probably should have added "high quality" to the prompt though
>>
>>101726865
I'm not against generating porn tho
>image limit reached
>>
:( I can't use Flux anymore today, its limiting me now for 8 mins. I only have
AMD Ryzen 7 5700G 57 °C
Cezanne 7nm Technology
RAM
48.0GB Dual-Channel DDR4 @ 1196MHz (16-16-16-39)
Motherboard
Gigabyte Technology Co., Ltd. AX370M-DS3H-CF (AM4) 41 °C
Graphics
12272MB ATI AMD Radeon RX 6750 XT (Unknown)
>>
>we hit image limit for once
BAKER!
>>
>>101726856
>A page taken from the Necronomicon written by Abdul Alhazred. The page is written in arabic along with various occult sigils and inscriptions in an unknown occult language. There are diagrams providing ritual instructions for communing with the Great Old Ones. The page appears authentic and ancient, and evokes a sense of indescribable dread.
>>
Baker baker baker man, bake me a thread as fast as you can.
I will not bake it not by myself,
someone else bake it and bake it top shelf.
>>
Baking.
>>
>>101726952
Thanks, Marmandy Jones.
>>
>>101726834
lmao I love this
>>
>>101726952
kys
>>
fuck that guy, I'm actually baking
>>
>>101726630
engaged in Gorilla warfare
>>
yo niggas who bakin?
>>
>>101727187
120s
>>
I'm too baked for this
>>
Not baking
>>
Is this how /ldg/ dies?
>>
These guys are trolling. I'm making the collage and will bake as soon as I finish.
>>
>>101727285
ty!
>>
Bump
>>
I'm baking, hold on!
>>
bake is up
>>
Need more bake-on
>>
Baker?
>>
https://www.youtube.com/watch?v=AeUwikDrA9I
>>
>>101727444
>>101727444
>>101727444

Emergency bake



[Advertise on 4chan]

Delete Post: [File Only] Style:
[Disable Mobile View / Use Desktop Site]

[Enable Mobile View / Use Mobile Site]

All trademarks and copyrights on this page are owned by their respective parties. Images uploaded are the responsibility of the Poster. Comments are owned by the Poster.