You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The result is a black output. Comfy terminal shows:
img = Image.fromarray(np.clip(i, 0, 255).astype(np.uint8))```
I tried fixing it various ways:
-appending --force-upcast-attention to run_gpu.bat (no luck)
-(I did notice the workflow above doesn't have the gguf CLIPLoader by default, so I added it, selecting the t5xxl_fp16-q4_0.gguf)
-I also tried using various safetensors in the default CLIPLoader
-I updated comfyUI but since the GGUF build seems custom, it didn't seem to update.
Please if you are reading this and you know the fix (or possible fix), please do share!! ;)
The text was updated successfully, but these errors were encountered:
you could get the most updated pack under releases; decompress it and try to run it again with the clean settings; for updating the node, can just clone this repo and replace the one inside ./ComfyUI/custom_nodes
Hi,
I am using your guide and comfy workflow as well as your ltx files and gguf comfyui
Followed these steps
run the .bat file in the main directory (assuming you are using the gguf-node pack below)
Download and place files:
-drag ltx-video-2b-v0.9.1-revised-q4_0.gguf (1.08GB) to > ./ComfyUI/models/diffusion_models
-drag t5xxl_fp16-q4_0.gguf (2.9GB) to > ./ComfyUI/models/text_encoders
-drag ltx-video-vae.safetensors (838MB) to > ./ComfyUI/models/vae
run it straight (no installation needed way)
However:
The result is a black output. Comfy terminal shows:
The text was updated successfully, but these errors were encountered: