วีดีโอ

can i add more images (front, back, left, right, top, bottom) for better result? thanks

You do realize that joining for free is different from €3 a month?

List object has no attribute shape

top man!

video thumbnail dreams, video content reality

I would like to know how to change the background. Thanks for this tutorial.

good simple and clear lesson!

FYI there is an update on cogvideox wrapper which removes the fun smapler need an update,

This node no longer works

how can fix KeyError: ((1, 1, 5), '|u1')??

Great, video, thank you.

Thanks man, really helpful, liked & subscribed

I am doing something wrog probably.I could not get workflow work.Could you plese share workflow too.Thank you

Compare to forge UI how much slower flux is in comfyui?

🥲my TripoSR view is not showing the 3d. It's show "loading scene...."

Someone any idea why i get an error message saying that IC Light expects 3 channel but my input image has 4? Tried everthing. Removed alpha channels, Changed base model, tested fbc and fc IC Light model. Nothing worked. My workflow does not include any other custom node.

Hey very nice video i have a task where i have a 2 car image and i want to replace 1 car image with 2 car image how can i do it ?

Nice tutorial, could you make a video on how to install everything from scratch?

Where is output folder

THANK YOU!!! I couldn't figure this out for days.

Dentro de que versión de Python necesita estar funcionando comfyUI para instalar estos nodos? Saludos.

Do I need to generate a Tripo API Key for this workflow?

what is the use of alpha channel as mask, any alternatives?

How can I make the character in a dance video perform at the original image size instead of a human size?

do not use sgm_uniform in sd

Thank you! Its posible to convert obj meshed to glb in comfyui? Thank you!

Thank you! Its posible to convert obj meshed to glb in comfyui? Thank you!

sooooooooo slowwwwwwwwww

wtf is this? thumbnail has nothing to do with the video/ clockbaiting in 2025?! seriously? this quality is crap

I saw your youtube looks great! I was wondering if you know of a Flux 1.1 dev node for loras (16 b)? (comfy UI) Happy you are showing great videos!

I correctly installed the deforum "nodes", however, I receive this message in comfyui "(IMPORT FAILED) Deforum Nodes"

It is November 6th and those nodes did the same thing to me where they just randomly connected to other nodes. I think it is something specific to KJNodes because it never happened until I installed them for your tutorial and it was the KJNodes that did it to you as well. Not a big deal of course but definitely a little annoying heheh

how do y get Tripo API key to start using it in Comfy ?

Isn't flowframes more easy and better?

Thanks for the comparison, very helpful! I guess they both have their advantages and disadvantages, but overall, on a stricly aesthetical level, I have to say I tend to prefer the SDXL outputs. For example, I found the statue faces too modern / photorealistic with 3.5, the colors overall too saturated, and, like you, the octopus very unconvincing. With regards to anatomy mistakes: On their civitai page, the SD people write: "We recommended to sample with Skip Layer Guidance for better struture and anatomy coherency." I downloaded the Skip Layer Guidance workflow they provide on their huggingface page, and I would agree that it definitely helps with anatomy, escpecially hands. (The workflow shows the output both with and without Skip Layer Guidance.)

Whether SD 3.5 will be in demand or not depends on ControlNet and A1111 support

what about comfyui 3d pack?

Thanks. Despite of anatomical flaws it generates, SD 3.5 feels quite interesting. While it can't quite generate vehicles, multiple characters, or distant characters yet (at least based on my tests), I think overall if community decides to do some fine-tunes, it could probably generate even higher quality images. And at least it can already generate nice looking faces, skin shading, nature scenes and unlike Flux Schnell, SD 3.5M is not a distilled model.

AlphaChanelAddByMask is just a big bug. Doesn't work.

dude how you get the triplecliploader?

It gives an error message. "AlphaChanelAddByMask too many values to unpack (expected 2)" :/ Do you know any solution, please?

Hi, I can't install or update GGUF custom nodes to get Unet Loader. I'm running Comfyui from Pinokio. Any help please ?

Unfortunately XmYx/deforum-comfy-nodes (IMPORT FAILED) and it seems that a few people on git have issues too. Even trying the "Try fix" in Comfy does not solve the problem. I was so excited about this video. Will have to wait for a fix. Nice video btw.

Thanks. Possible convert to 3d from images multi-view ?

I was looking around to find how to get rid of duplicated frames and found this. I have to say, you got the idea I need but implemented in an unnecessary complicated way. First of all, you don't need a complicated diagram to load 2 adjacent frames. Just set increment for them seperately starting with 0 and 1, and max value at last frame - 1 and last frame. That will solve the trick. Secondly, to get rid of duplicated frame, just split the result batchs and throw away last frame. After the whole process was done, manually add last frame. That's is alot easier and faster. Too much step will cost more computing power, which is a big deal when we only have a low-end machine.

Hi, super useful. I have this error: # ComfyUI Error Report ## Error Details - **Node Type:** CRMPoseSampler - **Exception Type:** NotImplementedError - **Exception Message:** No operator found for `memory_efficient_attention_forward` with inputs: query : shape=(1, 1024, 1, 512) (torch.float32) key : shape=(1, 1024, 1, 512) (torch.float32) value : shape=(1, 1024, 1, 512) (torch.float32) attn_bias : <class 'NoneType'> p : 0.0 `decoderF` is not supported because: max(query.shape[-1] != value.shape[-1]) > 128 xFormers wasn't build with CUDA support attn_bias type is <class 'NoneType'> operator wasn't built - see `python -m xformers.info` for more info `flshattF@0.0.0` is not supported because: max(query.shape[-1] != value.shape[-1]) > 256 xFormers wasn't build with CUDA support dtype=torch.float32 (supported: {torch.float16, torch.bfloat16}) operator wasn't built - see `python -m xformers.info` for more info `cutlassF` is not supported because: xFormers wasn't build with CUDA support operator wasn't built - see `python -m xformers.info` for more info `smallkF` is not supported because: max(query.shape[-1] != value.shape[-1]) > 32 xFormers wasn't build with CUDA support operator wasn't built - see `python -m xformers.info` for more info unsupported embed per head: 512 ## Stack Trace ```

Hi I cloned KJnodes into custom_nodes folder and did pip install -r requirements.txt and restarted the swarmui, but still I cannot get that grow mask with Blur option. Please help

This has been very useful in the past couple of months. I realised that if you have a video as input or made on the fly (like CogVideo5) you can just feed it into the interpolater and videcombine the output, and that just works without trouble. But I was making a flow that did something to the images and then wanted to interpolate that live, but couldn't work it out. So of course, I can make it save images out and then go through another stage like the above to make the interpolation. What happens is I get loads of tiny videos which are just each batch run of a single frame, what I wanted it to do was interpolate the image sequence and then save the video at the end without it creating the intermediate images.

"Value not in list: model: 'triposr.ckpt' not in (list of length 67) Output will be ignored" what are im doing wrong?

Thanks! Actually this one seems to be working a lot better than Florence 2 for certain things.