Hi, thanks for the video tutorial. When you press install missing custom nodes and install them, instead of pressing the red restart button you close the windows command prompt and you restart the command prompt ( 2:41 )AI Animation in ComfyUI with CogVideoX workflow, it is not visible how to do it?
Hey Vladimir, I am using a 4070 Nvidia card with 12 GB VRam, the video processing is incredibly slow already running 1 hour on a single inference step (1 out of 25). What kind of hardware are you using? I assume that this CogVideoX workflow requires a minimum hardware like a 4090 NVidia card with 24 GB VRam, is that correct?
@@Geekatplay Thanks, something must be wrong with my ComfyUI configuration. How long does it take you to render the "Ice queen" video on your 4060? (I believe that card has 8 GB Vram)
annoying things like download everything then.. queue and it start download correct ones that is forced to use. not ones on link, but it also right way. only need workflow and hit queue and it download if not have. then it load it memory. takes long time first test and know it sure not going to be what expect. result is what is trained. like i allways say if there no banana AI not do banana. it wont generate nothing from thin air
22:15 did what trained but its wrong.. its trained data face outline. not this person. now try image with person that have nose broken. it wont do it. it try do nose like trained
@@Geekatplay not matter havent done anything yet. not enought memory this machine. just downloading all models before start better machine and move whole comfuui folders there XD
![Vladimir Chopine [GeekatPlay]](http://yt3.ggpht.com/C8QIoKOqO3xNbQhrLw7Q48NVh8ynPa99GNZ2qYvEN4Csg_nwKRnLcMFYfEdljEseP84ZCfUixQ=s48-c-k-c0x00ffffff-no-rj)
Hi, thanks for the video tutorial. When you press install missing custom nodes and install them, instead of pressing the red restart button you close the windows command prompt and you restart the command prompt ( 2:41 )AI Animation in ComfyUI with CogVideoX workflow, it is not visible how to do it?
what do you mean not visible?
@@Geekatplay That we can't see how you restart the command prompt, by pressing the red restart button, closing comfyui and reopening it?
Hey Vladimir, I am using a 4070 Nvidia card with 12 GB VRam, the video processing is incredibly slow already running 1 hour on a single inference step (1 out of 25). What kind of hardware are you using? I assume that this CogVideoX workflow requires a minimum hardware like a 4090 NVidia card with 24 GB VRam, is that correct?
check what else running on your computer. My main PC RTX3090, however i run also on RTX4060 and it is working ok
@@Geekatplay Thanks, something must be wrong with my ComfyUI configuration. How long does it take you to render the "Ice queen" video on your 4060? (I believe that card has 8 GB Vram)
And if I may ask: what is the version of Python you are using for this Workflow?
annoying things like download everything then.. queue and it start download correct ones that is forced to use. not ones on link, but it also right way. only need workflow and hit queue and it download if not have. then it load it memory. takes long time first test and know it sure not going to be what expect. result is what is trained. like i allways say if there no banana AI not do banana. it wont generate nothing from thin air
22:15 did what trained but its wrong.. its trained data face outline. not this person. now try image with person that have nose broken. it wont do it. it try do nose like trained
try with solid background, it is using depth map to extract face
@@Geekatplay not matter havent done anything yet. not enought memory this machine. just downloading all models before start better machine and move whole comfuui folders there XD