Great job! I'm testing this and it works wonders. I have never been able to get that upscale to work as good as yours. Thank you for sharing this workflow.
I'm with you man. Tryin' whatever may come. I don't have the courage to make vidoes but i like your ideas and us all figuring this stuff out. Keep up the good work . thanks
This is awesome. Thank you for making and sharing. In your experiments, have you gained insights as to when would you prefer to generate using SDXL Turbo vs LCM ?
Nice video giving back to the community! For future videos you should consider turning off the music off after your introduction. Makes it hard to hear you.
You have made an excellent job on this one. Gonna try as soon as I can, so far the workflows for turbo aren't as sophisticated as this like the idea of reselecting a result and then following to the next operation. I was making my own superworkflow mega spaghetti but I made a mistake of creating all my logic before testing and when I started testing realised the ComfyUI doesn't follow my switches logic and rather counts the nodes that are nearer to the result I am still super confused why but have no time to sit and study it yet😅.
Thanks for the nice comment. ImpactPack's any switch node is so powerful, I did notice that for some types though if I want to connect the output of the switch to a reroute, it doesn't work with the default comfy one, but it does work with using the first reroute node from the pythongosss pack, then after the default comfy reroutes can have that as an input. I'm not familiar with how the logic behind the scenes fully works, which is why I've done my flows in sections. This way users can use or not use parts of it whenever they want.
@@boricuapabaiartist Thanks I will check Impact Pack yeah for now I am breaking my workflow and discard the parts that are not working separatly until I find a solution you should check the openart contst I am gonna try to enter too :D They looking for workflow submitions
Hey there, I came her from the comfy workflows page as I was looking for your video for the Carnival Dancer, it is currently set to private. Any chance you can reupload/relist/whatever that video again? I would love to have a look at how your workflow is used as I am basically new to comfy.
sorry probably a simple question but to create new seeds , is the only way to manually change or can you create new seeds that are immediately fixed for taking onto the next stages?
I know this is late but "prevLtntNSeedChooser" in the selectLatentImageToSave section is highlighted red and not working, I've installed all missing nodes, any idea?
I've updated the workflow as the Derfuu custom node pack doesn't have an integer node now in the latest version, so I changed it to the integer node from the Comfy Literals custom node pack, I've re-uploaded a version 1.2 of the workflow, links in the description
Your workflow is great but for some reason I can't figure out, my previews are super small. They maintain the aspect ratio but the smaller lenght or height is 64px.
Check your comfyui models/vae_approx folder you should have a file called taesdxl_decoder.pth to get the full resolution preview on the preview latent node, it usually auto-downloads into that folder, or you can download it from the comfyUI repo: github.com/comfyanonymous/ComfyUI scroll all the way to the bottom in the section titled: How to show high-quality previews?
Hello, It looks cool but I tried my prompt "Scenes of happy customers receiving packages, opening products with smiles, and leaving positive reviews." Faces are deformed :(
Yea, some diffusion models aren't too good at making scenes with multiple characters in it with nice faces. You'd need to upscale the image to a higher resolution and inpaint the faces with a denoise value from around .52 to .6 to make them better. You can also try the face detailer node. I'd need to expand upon this workflow to add the ability for doing those 2 things. And possibly for the inpaint step and new k sampler it would be better to use non turbo models to get the higher quality results.
>>>> I believe we are meant to be like Jesus in our hearts and not in our flesh. But be careful of AI, for it is just our flesh and that is it. It knows only things of the flesh (our fleshly desires) and cannot comprehend things of the spirit such as peace of heart (which comes from obeying God's Word). Whereas we are a spirit and we have a soul but live in the body (in the flesh). When you go to bed it is your flesh that sleeps but your spirit never sleeps (otherwise you have died physically) that is why you have dreams. More so, true love that endures and last is a thing of the heart (when I say 'heart', I mean 'spirit'). But fake love, pretentious love, love with expectations, love for classic reasons, love for material reasons and love for selfish reasons that is a thing of our flesh. In the beginning God said let us make man in our own image, according to our likeness. Take note, God is Spirit and God is Love. As Love He is the source of it. We also know that God is Omnipotent, for He creates out of nothing and He has no beginning and has no end. That means, our love is but a shadow of God's Love. True love looks around to see who is in need of your help, your smile, your possessions, your money, your strength, your quality time. Love forgives and forgets. Love wants for others what it wants for itself. Take note, true love works in conjunction with other spiritual forces such as patience and faith (in the finished work of our Lord and Savior, Jesus Christ, rather than in what man has done such as science, technology and organizations which won't last forever). To avoid sin and error which leads to the death of our body and also our spirit in hell fire, we should let the Word of God be the standard of our lives not AI. If not, God will let us face AI on our own and it will cast the truth down to the ground, it will be the cause of so much destruction like never seen before, it will deceive many and take many captive in order to enslave them into worshipping it and abiding in lawlessness. We can only destroy ourselves but with God all things are possible. God knows us better because He is our Creater and He knows our beginning and our end. Our prove text is taken from the book of John 5:31-44, 2 Thessalonians 2:1-12, Daniel 7-9, Revelation 13-15, Matthew 24-25 and Luke 21. Let us watch and pray... God bless you as you share this message to others.
thanks for sharing but even more important thanks for going through everything step by step
Great workflow - i changed the seed nodes to the rgthree seed node which allows you to create a new random fixed seed
amazing; you are a Comfy god!
Great job! I'm testing this and it works wonders. I have never been able to get that upscale to work as good as yours. Thank you for sharing this workflow.
I'm with you man. Tryin' whatever may come. I don't have the courage to make vidoes but i like your ideas and us all figuring this stuff out. Keep up the good work . thanks
Thanks, yes the learning aspect of all this is part of the fun. There's always something new to try.
Nice dude, first time bumping in to your channel, will give it a go and give that 4090 a nice workout.
Awesome, thanks for watching the vid and trying out my flow
This is awesome. Thank you for making and sharing. In your experiments, have you gained insights as to when would you prefer to generate using SDXL Turbo vs LCM ?
Nice video giving back to the community! For future videos you should consider turning off the music off after your introduction. Makes it hard to hear you.
Thanks for the feedback, I'll make sure that any background music I add onto my videos don't compete against my voice in future videos.
WELL DONE ! Thank you for sharing
awesome
Great workflow
Thanks, I enjoyed making this one.
You have made an excellent job on this one. Gonna try as soon as I can, so far the workflows for turbo aren't as sophisticated as this like the idea of reselecting a result and then following to the next operation.
I was making my own superworkflow mega spaghetti but I made a mistake of creating all my logic before testing and when I started testing realised the ComfyUI doesn't follow my switches logic and rather counts the nodes that are nearer to the result I am still super confused why but have no time to sit and study it yet😅.
Thanks for the nice comment. ImpactPack's any switch node is so powerful, I did notice that for some types though if I want to connect the output of the switch to a reroute, it doesn't work with the default comfy one, but it does work with using the first reroute node from the pythongosss pack, then after the default comfy reroutes can have that as an input. I'm not familiar with how the logic behind the scenes fully works, which is why I've done my flows in sections. This way users can use or not use parts of it whenever they want.
@@boricuapabaiartist Thanks I will check Impact Pack yeah for now I am breaking my workflow and discard the parts that are not working separatly until I find a solution you should check the openart contst I am gonna try to enter too :D
They looking for workflow submitions
Hey there, I came her from the comfy workflows page as I was looking for your video for the Carnival Dancer, it is currently set to private. Any chance you can reupload/relist/whatever that video again? I would love to have a look at how your workflow is used as I am basically new to comfy.
Hey man that's a great workflow and video. Thanks! Liked, commented and subscribed (already earlier). The least I can do! Keep it up man 👌
Thanks for checking out my videos and using the workflows.
sorry probably a simple question but to create new seeds , is the only way to manually change or can you create new seeds that are immediately fixed for taking onto the next stages?
I know this is late but "prevLtntNSeedChooser" in the selectLatentImageToSave section is highlighted red and not working, I've installed all missing nodes, any idea?
I've updated the workflow as the Derfuu custom node pack doesn't have an integer node now in the latest version, so I changed it to the integer node from the Comfy Literals custom node pack, I've re-uploaded a version 1.2 of the workflow, links in the description
Your workflow is great but for some reason I can't figure out, my previews are super small. They maintain the aspect ratio but the smaller lenght or height is 64px.
Check your comfyui models/vae_approx folder you should have a file called taesdxl_decoder.pth to get the full resolution preview on the preview latent node, it usually auto-downloads into that folder, or you can download it from the comfyUI repo:
github.com/comfyanonymous/ComfyUI
scroll all the way to the bottom in the section titled:
How to show high-quality previews?
that's awesome, I'm gonna try that on chicks models, you rock
I am trying to combine this wf with multiarea conditiong but no success yet. I will be digging hardly tn. Any contribution welcoming guys 🙏🏿
I haven't used multi area conditioning much apart from the default examples workflow comfy has up on the comfyui repo
Hello, It looks cool but I tried my prompt "Scenes of happy customers receiving packages, opening products with smiles, and leaving positive reviews." Faces are deformed :(
Yea, some diffusion models aren't too good at making scenes with multiple characters in it with nice faces. You'd need to upscale the image to a higher resolution and inpaint the faces with a denoise value from around .52 to .6 to make them better. You can also try the face detailer node. I'd need to expand upon this workflow to add the ability for doing those 2 things. And possibly for the inpaint step and new k sampler it would be better to use non turbo models to get the higher quality results.
@@boricuapabaiartist Thank you for advice, I will do it
the music is too loud ;)
>>>> I believe we are meant to be like Jesus in our hearts and not in our flesh. But be careful of AI, for it is just our flesh and that is it. It knows only things of the flesh (our fleshly desires) and cannot comprehend things of the spirit such as peace of heart (which comes from obeying God's Word). Whereas we are a spirit and we have a soul but live in the body (in the flesh). When you go to bed it is your flesh that sleeps but your spirit never sleeps (otherwise you have died physically) that is why you have dreams. More so, true love that endures and last is a thing of the heart (when I say 'heart', I mean 'spirit'). But fake love, pretentious love, love with expectations, love for classic reasons, love for material reasons and love for selfish reasons that is a thing of our flesh. In the beginning God said let us make man in our own image, according to our likeness. Take note, God is Spirit and God is Love. As Love He is the source of it. We also know that God is Omnipotent, for He creates out of nothing and He has no beginning and has no end. That means, our love is but a shadow of God's Love. True love looks around to see who is in need of your help, your smile, your possessions, your money, your strength, your quality time. Love forgives and forgets. Love wants for others what it wants for itself. Take note, true love works in conjunction with other spiritual forces such as patience and faith (in the finished work of our Lord and Savior, Jesus Christ, rather than in what man has done such as science, technology and organizations which won't last forever). To avoid sin and error which leads to the death of our body and also our spirit in hell fire, we should let the Word of God be the standard of our lives not AI. If not, God will let us face AI on our own and it will cast the truth down to the ground, it will be the cause of so much destruction like never seen before, it will deceive many and take many captive in order to enslave them into worshipping it and abiding in lawlessness. We can only destroy ourselves but with God all things are possible. God knows us better because He is our Creater and He knows our beginning and our end. Our prove text is taken from the book of John 5:31-44, 2 Thessalonians 2:1-12, Daniel 7-9, Revelation 13-15, Matthew 24-25 and Luke 21. Let us watch and pray... God bless you as you share this message to others.