FLux.1 Local Install Made Easy - The BEST FREE UNCENSORED AI Model Now!
Вставка
- Опубліковано 20 вер 2024
- In this video, I introduce Flux, a powerful new text-to-image AI model from Black Forest Labs, built by former Stability AI developers. Flux surpasses Stable Diffusion 3, offering incredible prompt accuracy, stunning full HD outputs, and more. I'll guide you through trying it online, installing it locally, configuring prompts, and upscaling images. Whether you're using the Schnell, Dev, or Pro version, I'll share tips to optimize performance and get the most out of this game-changing tool. Ready to elevate your AI-generated images? Let’s dive in!
Book a one-on-one appointment: koalendar.com/...
Patreon Free Stuff: / aiknowledgecentral
Ultimate Beginners Guide: / sneak-peek-alert-90799508
Free workflow guide: / get-your-free-99183367
Style Collection: / my-collection-of-87325880
My ComfyUI Tuts:
• ComfyUI
My Automatic 1111 Tuts:
• Automatic 1111 - Stabl...
AI Tools:
• AI-Tools
Resources:
Flux.1 Schnell online:
huggingface.co...
Flux full
huggingface.co...
VAE:
huggingface.co...
Text encoders
huggingface.co...
Workflow:
comfyanonymous...
SDXL:
civitai.com/mo...
VAE:
huggingface.co...
Flux fp8 - last resort fallback
huggingface.co...
Gerold Mesinger Sampler -Scheduler test:
/ fluxdev_samplers_and_s...
You can find the prompts and the workflow on my patreon:
www.patreon.com/AIKnowledgeCentral
Do you struggle with AI-Art? Head over to my Patreon to grab free stuff. No Membership needed!
Ultimate Beginners Guide: www.patreon.com/posts/sneak-peek-alert-90799508
Free workflow guide: www.patreon.com/posts/get-your-free-99183367
Style Collection: www.patreon.com/posts/my-collection-of-87325880
I also offer one on one sessions:
koalendar.com/e/1hr-1-on-1-stable-diffusion
Happy creating
Chris
Good video. Please make another video on how to integrate "img to img" using flux. This will help us take existing image and write text on top of it instead of generating an image with text from scrach.
Thanks for your feedback! I'll definitely make a video about this in the future.
@@AIKnowledge2Go ok waiting :)
image flashing at 30sec is mind melting
Thanks for watching! Yeah I sometime overdo it with the effects and my pacing is off.
@@AIKnowledge2Go i meant the image that you flash there is mind meltingly epic!
Thanks for video. 👍
I'm glad you enjoyed the video! Your support means a lot to me.
[5:33] "Oh, how did that get in here?! I have no idea."
- LOL! (literally) :D
- Well played
Haha, glad you enjoyed that moment! It definitely caught me off guard too when i wrote it 😂!
@@AIKnowledge2Go
- Nice explanation, BTW.
- Thx for your effort.
- I started in computer graphics, and 3D 30+ years ago, and the progress, especially recent, has been amazing and fun to watch (and be part of: as a modeler/animator, and developer of custom code for 3D).
- Keep up the great content!...
Thanks a lot! As usual very clear and useful. What SDXL checkpoint are you using for the upscaling? My image comes out blurred. I might be doing something wrong. I am also using the FLux1-dev maybe that could be the problem. Thanks again for the great content.
Thanks for the comment i appreciate it. I am using Cyberrealistic XL 2.1 i don't think its the dev version. What Resolution are you using, and to what resolution are you upscaling? Can you check that your settings resemble mine in terms of Steps, Sampler etc.?
@AIKnowledge2Go I managed to upscale it. I set de denoise to 0.50 - 0.60, then I used juggernautXL. My resolution is 9:16 and updcaling by 1.50.
Funny how sd3 is done when I never even moved from sd1.5
I terms of tooling, SD 1.5 is still the best! The SDXL ControlNet Models for example are sh*t. Inpainting, outpainting etc. just worked in 1.5. Happy creating.
Why does the saftensor not go in the checkpoint folder, but in the UNET
Ask the guys at Stabillity AI or ComfyUI developers (Maybe they changed it in the meantime, i don't check small issues). Happy creating.
plus one minute for just one image and only one pass is not worth it, it's not worth the power it draws, the heat it generates and the time it waste. it's inefficient at best, GUUF fix v-ram issues but the generation is even slower. we need good model that can fit on lower vram machines and generates faster than sdxl lightning, other wise it's a waste of resources.
I feel you, currently we are somewhat stuck with the ai models...
@@AIKnowledge2Go Yep :(
It may be uncensored but it has no clue who many famous people are. Try to get it to make you a Barbara Eden, for example
Its not really uncensored either. Not "out of the box" at least
The AI image model likely struggles with certain famous people because their likenesses were deliberately excluded from the training data to avoid legal issues. Why this particularly affects female figures, I can't say.
We need to differentiate: some models/text encoders are censored and will filter out certain NSFW prompts. Flux, on the other hand, isn't censored but simply hasn't been trained on certain data yet.
That’s a good thing, because some dummies would make stupid shit bringing new laws.
@@AIKnowledge2Go I tried naked woman on grass and the n@pples where somehow blurbed in red. I think it is clearly censored
you Dutch past simple cant go away with english lol
Thanks for the feedback! I'm actually German, not Dutch, but I'm always working on improving my English. Feel free to point out any mistakes-I appreciate the help!
@@AIKnowledge2Go Your English is better than the commenter's my friend. Ignore them. Their sentence was gibberish; yours was perfectly written and punctuated.
@@clarkkent4734 Thank you for your support. 😊
@@AIKnowledge2GoIch wusste, dass du deutsch bist, als Du Flax statt flux gesagt hast, wie auf dem englischen Kanälen
As soon as i hit the generate button my ram starts to fill and and then get to the max. I don't get any generation. Everything is checked and i didn't get any errors and i also renamed sft to safetensors. I have rtx4090, how can i fix it?
I am stuck here.
got prompt
model weight dtype torch.bfloat16, manual cast: None
model_type FLOW
Can you use ComfyUI with any other workflow, lets say a basic SD 1.5 or SDXL workflow? I don't know much about your system so i may be wrong here, but to me that sounds like your system tries to render using the cpu or an onboard gpu.
@@AIKnowledge2Go my cpu doesn't take too much load during the process and gpu takes around 15% and my ram (32gb) starts to fill up and goes upto 31.6 and remains there until i closs the dos window to restart the comfyui because of too much time passing.
Yes i am using and can use sd 1.5 and sdxl but on my flux workflow i am only using flux. Only flux is not working for me. I tried both full model and the schnell model and getting same issue.
My system is
I9 13900k
Rtx4090
32gb 6600MT ddr5 ram
@@zaselimgamingvideos To be honest, at this point, it feels a bit like we're grasping at straws, but let’s try a few more things to diagnose the issue:
Have you tried setting your model weight to FP8_e4m3fn and configuring your NVIDIA card to prefer no system fallback? Sometimes these settings can influence how resources are managed. Can you confirm if CUDA 1.21 is installed on your system? Having the right version can be crucial for compatibility and performance.
Additionally, a fresh reinstall of ComfyUI might help clear up any configuration issues or corruptions that might have occurred during the initial installation or during use.
@@AIKnowledge2Go yes my nvidia is previously set to no fall back. If i remember correctly my cuda says 12.something when i load automatic 1111 or comfyui.
Actually it is a fresh install, i did after your and couple other videos. Before that i was using A1111. I installed comfyui to try out flux. And yes i tried both default and fp8_e4m3fn just to check if comfy is not registering my vram as 24gb. But it does when i launch comfy, bram 24gb appears on the doss window. I will check confirm the cuda version on comfy but i am 100% its 12.something on my a1111 after the recent update.
@@AIKnowledge2Go i just checked my cuda version by command nvidia-smi and it says i have cuda 12.5