This works for me right up to 6:55 but when I hit the Queue run button, I get the following: "Prompt outputs failed validation CLIPLoader: - Value not in list: clip_name: 't5\google_t5-v1_1-xxl_encoderonly-fp8_e4m3fn.safetensors' not in [] DownloadAndLoadMochiModel: - Value not in list: vae: 'mochi_preview_vae_bf16.safetensors' not in ['mochi_preview_vae_decoder_bf16.safetensors']" Any help is appreciated.
you need to set the VAE model. You can download it from huggingface.co/mcmonkey/google_t5-v1_1-xxl_encoderonly/blob/main/t5xxl_fp8_e4m3fn.safetensors and then put it in comfyui/models/vae directory. Thanks.
@@fahdmirza Thank you for the blisteringly fast reply. I was able to go to the Model Manager, search for "t5/t5" and find "google-t5/t5-v1_1-xxl_encoderonly-fp8_e4m3fn" which I downloaded. Then I was still stuck on the second part "Value not in list: vae: 'mochi_preview_vae_bf16.safetensors' not in ['mochi_preview_vae_decoder_bf16.safetensors']" For that one, I was able to randomly click on the Download Mochi Model like of "vae....." and it corrected itself [I saw this in another video] All good now. Thank you for your help!
Didn't download the models Prompt outputs failed validation CLIPLoader: - Value not in list: clip_name: 't5\google_t5-v1_1-xxl_encoderonly-fp8_e4m3fn.safetensors' not in [] DownloadAndLoadMochiModel: - Value not in list: vae: 'mochi_preview_vae_bf16.safetensors' not in ['mochi_preview_vae_decoder_bf16.safetensors']
trying this I found on a Civitiai page .. t5xxl_fp8_e4m3fn.safetensors - Search for "t5xxl_fp8" in the Model Manager if the full name doesn’t appear. huggingface comfyanonymous/flux_text_encoders/tree/main
only needs about 20gb VRAM , runs fine on 3090 24gb vram , or 4090 . using 3090 takes one time 5 min to first load the model , then after that about 7min to generate 3 second video at 848x480 resolution
the easiest and still freaking hard. I wonder when will they release a comfy for non-coders. when loading the workflow I still have Missing nodes. missing nodes don't appear when clicking install missing nodes. missing nodes are MochiTextEncoder, Mochisampler, MochiDecode and there's another one. can't remember.
It is Helpful, but only up to the stage where you opened ComfyUi and lost me. there was explaination on how to get the console up lol. I installed git that was what I was missing.
@@fahdmirza Where it says Load Clip, and clip name, it says Null. when i hit prompt its comes up with ``Prompt outputs failed validation CLIPLoader: - Value not in list: clip_name: 'None' not in []`` this is where im lost. i cannot work out what to do lol.
will it run on System with RTX 3050 gpu with 4 gb vram and Ram 24 GB,can the GGUF Q4 models run atleast,Please tell...Is there a image to video version for low vram systems to run locally.
Prompt outputs failed validation CLIPLoader: - Value not in list: clip_name: 't5\google_t5-v1_1-xxl_encoderonly-fp8_e4m3fn.safetensors' not in ['clip_l.safetensors', 't5xxl_fp16.safetensors', 't5xxl_fp8_e4m3fn.safetensors', 't5xxl_fp8_e4m3fn_scaled.safetensors'] DownloadAndLoadMochiModel: - Value not in list: vae: 'mochi_preview_vae_bf16.safetensors' not in ['mochi_preview_vae_decoder_bf16.safetensors'] ???
![BLACK BAG - Official Trailer [HD] - Only in Theaters March 14](http://i.ytimg.com/vi/Du0Xp8WX_7I/mqdefault.jpg)
This works for me right up to 6:55 but when I hit the Queue run button, I get the following: "Prompt outputs failed validation
CLIPLoader:
- Value not in list: clip_name: 't5\google_t5-v1_1-xxl_encoderonly-fp8_e4m3fn.safetensors' not in []
DownloadAndLoadMochiModel:
- Value not in list: vae: 'mochi_preview_vae_bf16.safetensors' not in ['mochi_preview_vae_decoder_bf16.safetensors']"
Any help is appreciated.
you need to set the VAE model. You can download it from huggingface.co/mcmonkey/google_t5-v1_1-xxl_encoderonly/blob/main/t5xxl_fp8_e4m3fn.safetensors and then put it in comfyui/models/vae directory. Thanks.
@@fahdmirza Thank you for the blisteringly fast reply. I was able to go to the Model Manager, search for "t5/t5" and find "google-t5/t5-v1_1-xxl_encoderonly-fp8_e4m3fn" which I downloaded. Then I was still stuck on the second part "Value not in list: vae: 'mochi_preview_vae_bf16.safetensors' not in ['mochi_preview_vae_decoder_bf16.safetensors']" For that one, I was able to randomly click on the Download Mochi Model like of "vae....." and it corrected itself [I saw this in another video]
All good now. Thank you for your help!
i dont have this window on the right (Load, save etc)
Upgrade your ComfyUI
@@fahdmirza ComfyUI is already up to date with the latest version.
Didn't download the models
Prompt outputs failed validation
CLIPLoader:
- Value not in list: clip_name: 't5\google_t5-v1_1-xxl_encoderonly-fp8_e4m3fn.safetensors' not in []
DownloadAndLoadMochiModel:
- Value not in list: vae: 'mochi_preview_vae_bf16.safetensors' not in ['mochi_preview_vae_decoder_bf16.safetensors']
trying this I found on a Civitiai page ..
t5xxl_fp8_e4m3fn.safetensors - Search for "t5xxl_fp8" in the Model Manager if the full name doesn’t appear.
huggingface comfyanonymous/flux_text_encoders/tree/main
have you got the solution ?
@SagarVerma16 Yes I ended up just installing the one on Pinocchio there's a mochi on there
@@edphonez what does this mean
@@edphonez how do you do ? help me pls
may I ask what’s your gpu, it’s vram and ram in your cpu?
Its mentioned in video
Do we still need 4 h100 gpu
only needs about 20gb VRAM , runs fine on 3090 24gb vram , or 4090 . using 3090 takes one time 5 min to first load the model , then after that about 7min to generate 3 second video at 848x480 resolution
the easiest and still freaking hard. I wonder when will they release a comfy for non-coders. when loading the workflow I still have Missing nodes. missing nodes don't appear when clicking install missing nodes. missing nodes are MochiTextEncoder, Mochisampler, MochiDecode and there's another one. can't remember.
Try to use ComfyUI Manager to install missing nodes.
@@fahdmirza I wrote clearly that didn't work bro.
@@cj5787search independently
It is Helpful, but only up to the stage where you opened ComfyUi and lost me. there was explaination on how to get the console up lol. I installed git that was what I was missing.
Sure, plz watch the comfyui tutorial which I mentioned in video.
@@fahdmirza Where it says Load Clip, and clip name, it says Null. when i hit prompt its comes up with ``Prompt outputs failed validation
CLIPLoader:
- Value not in list: clip_name: 'None' not in []`` this is where im lost. i cannot work out what to do lol.
@@MrOptimusheath Same issue
@@cloudu0 if you find a solution please let me know
will it run on System with RTX 3050 gpu with 4 gb vram and Ram 24 GB,can the GGUF Q4 models run atleast,Please tell...Is there a image to video version for low vram systems to run locally.
Prompt outputs failed validation
CLIPLoader:
- Value not in list: clip_name: 't5\google_t5-v1_1-xxl_encoderonly-fp8_e4m3fn.safetensors' not in ['clip_l.safetensors', 't5xxl_fp16.safetensors', 't5xxl_fp8_e4m3fn.safetensors', 't5xxl_fp8_e4m3fn_scaled.safetensors']
DownloadAndLoadMochiModel:
- Value not in list: vae: 'mochi_preview_vae_bf16.safetensors' not in ['mochi_preview_vae_decoder_bf16.safetensors']
???
Do u remember me 😊😊😊
yes