Lightricks
no CLIP/text encoder weights in checkpoint, the text encoder model will not be loaded.
Try to run ltxv-13b-0.9.7-distilled-fp8, on my windows 11, 4080s, there always stop at 44% and return "no CLIP/text encoder weights in checkpoint, the text encoder model will not be loaded."
but i have put t5xxl_fp16 in the text_endcoders folder.
To see the GUI go to: http://127.0.0.1:8188
FETCH ComfyRegistry Data: 5/90
FETCH ComfyRegistry Data: 10/90
FETCH ComfyRegistry Data: 15/90
FETCH ComfyRegistry Data: 20/90
got prompt
model weight dtype torch.float8_e4m3fn, manual cast: torch.bfloat16
model_type FLUX
FETCH ComfyRegistry Data: 25/90
FETCH ComfyRegistry Data: 30/90
VAE load device: cuda:0, offload device: cpu, dtype: torch.bfloat16
no CLIP/text encoder weights in checkpoint, the text encoder model will not be loaded.****
ComfyUI was updated to date ComfyUI: v0.3.43 (2025-06-27) Manager: V3.33.4
It is normal behavior for the model to only include the model itself and the VAE. The CLIP needs to be loaded separately. Typically, the google/t5-v1_1-xxl works well and is available in multiple formats, such as GGUF.
If I remember correctly, this message:
No CLIP/text encoder weights in checkpoint; the text encoder model will not be loaded.
only occurs when the model is loaded from the node, but you are supposed to load the CLIP model with a separate node. This message is normal.
