ver 1.5.1 Error with sd3_medium_incl_clips_t5xxlfp16.safetensors
ptits opened this issue · 3 comments
it runs well with sd3_medium_incl_clips.safetensors
but with sd3_medium_incl_clips_t5xxlfp16.safetensors it gives errors:
python entry_with_update.py --share
You have the latest version
/data/ptits/focus/RuinedDS3/RuinedFooocus/venv/lib/python3.10/site-packages/huggingface_hub/file_download.py:1132: FutureWarning: `resume_download` is deprecated and will be removed in version 1.0.0. Downloads always resume when possible. If you want to force a new download, use `force_download=True`.
warnings.warn(
Python 3.10.12 (main, Nov 20 2023, 15:14:05) [GCC 11.4.0]
RuinedFooocus version: 1.51.0
Comfy Backend update check complete.
playsound is relying on another python subprocess. Please use `pip install pygobject` if you want playsound to run more efficiently.
Total VRAM 81051 MB, total RAM 1290100 MB
pytorch version: 2.3.1+cu121
WARNING[XFORMERS]: xFormers can't load C++/CUDA extensions. xFormers was built for:
PyTorch 2.1.1+cu121 with CUDA 1201 (you have 2.3.1+cu121)
Python 3.10.13 (you have 3.10.12)
Please reinstall xformers (see https://github.com/facebookresearch/xformers#installing-xformers)
Memory-efficient attention, SwiGLU, sparse and more won't be available.
Set XFORMERS_MORE_DETAILS=1 for more details
xformers version: 0.0.23
Set vram state to: NORMAL_VRAM
Device: cuda:0 NVIDIA A100-SXM4-80GB : native
VAE dtype: torch.bfloat16
Using pytorch cross attention
Running on local URL: http://127.0.0.1:7860
Loading base model: sd_xl_base_1.0_0.9vae.safetensors
IMPORTANT: You are using gradio version 3.50.2, however version 4.29.0 is available, please upgrade.
--------
Running on public URL: https://9184c63067f4becfda.gradio.live
This share link expires in 72 hours. For free permanent hosting and GPU upgrades, run `gradio deploy` from Terminal to deploy to Spaces (https://huggingface.co/spaces)
model_type EPS
Using pytorch attention in VAE
Using pytorch attention in VAE
Base model loaded: sd_xl_base_1.0_0.9vae.safetensors
App started successful. Use the app with http://127.0.0.1:7860/ or 127.0.0.1:7860 or https://9184c63067f4becfda.gradio.live
Loading base model: sd3_medium_incl_clips_t5xxlfp16.safetensors
model_type FLOW
Using pytorch attention in VAE
Using pytorch attention in VAE
Base model loaded: sd3_medium_incl_clips_t5xxlfp16.safetensors
LoRAs loaded: []
Requested to load SD3ClipModel_
Loading 1 new model
Time taken: 1.97 seconds Pipeline process
Exception in thread Thread-5 (worker):
Traceback (most recent call last):
File "/usr/lib/python3.10/threading.py", line 1016, in _bootstrap_inner
self.run()
File "/usr/lib/python3.10/threading.py", line 953, in run
self._target(*self._args, **self._kwargs)
File "/data/ptits/focus/RuinedDS3/RuinedFooocus/modules/async_worker.py", line 341, in worker
handler(task)
File "/data/ptits/focus/RuinedDS3/RuinedFooocus/modules/async_worker.py", line 57, in handler
process(gen_data)
File "/data/ptits/focus/RuinedDS3/RuinedFooocus/modules/async_worker.py", line 240, in process
imgs = pipeline.process(
File "/data/ptits/focus/RuinedDS3/RuinedFooocus/venv/lib/python3.10/site-packages/torch/utils/_contextlib.py", line 115, in decorate_context
return func(*args, **kwargs)
File "/data/ptits/focus/RuinedDS3/RuinedFooocus/modules/sdxl_pipeline.py", line 280, in process
if self.textencode("+", positive_prompt, clip_skip):
File "/data/ptits/focus/RuinedDS3/RuinedFooocus/modules/sdxl_pipeline.py", line 202, in textencode
self.conditions[id]["cache"] = CLIPTextEncode().encode(
File "/data/ptits/focus/RuinedDS3/RuinedFooocus/repositories/ComfyUI-from-StabilityAI-Official/nodes.py", line 58, in encode
cond, pooled = clip.encode_from_tokens(tokens, return_pooled=True)
File "/data/ptits/focus/RuinedDS3/RuinedFooocus/repositories/ComfyUI-from-StabilityAI-Official/comfy/sd.py", line 142, in encode_from_tokens
cond, pooled = self.cond_stage_model.encode_token_weights(tokens)
File "/data/ptits/focus/RuinedDS3/RuinedFooocus/repositories/ComfyUI-from-StabilityAI-Official/comfy/sd3_clip.py", line 124, in encode_token_weights
t5_out, t5_pooled = self.t5xxl.encode_token_weights(token_weight_pars_t5)
File "/data/ptits/focus/RuinedDS3/RuinedFooocus/repositories/ComfyUI-from-StabilityAI-Official/comfy/sd1_clip.py", line 40, in encode_token_weights
out, pooled = self.encode(to_encode)
File "/data/ptits/focus/RuinedDS3/RuinedFooocus/repositories/ComfyUI-from-StabilityAI-Official/comfy/sd1_clip.py", line 201, in encode
return self(tokens)
File "/data/ptits/focus/RuinedDS3/RuinedFooocus/venv/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "/data/ptits/focus/RuinedDS3/RuinedFooocus/venv/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl
return forward_call(*args, **kwargs)
File "/data/ptits/focus/RuinedDS3/RuinedFooocus/repositories/ComfyUI-from-StabilityAI-Official/comfy/sd1_clip.py", line 186, in forward
z = outputs[1].float()
AttributeError: 'NoneType' object has no attribute 'float'
I just reproduced it on Windows 10, not only on Ubuntu as above
version 1.5.1, launched with update
========================
Exception in thread Thread-8 (worker):
Traceback (most recent call last):
File "threading.py", line 1016, in _bootstrap_inner
File "threading.py", line 953, in run
File "F:\RuinedFooocus_win64_1-25-2\RuinedFooocus\modules\async_worker.py", line 341, in worker
handler(task)
File "F:\RuinedFooocus_win64_1-25-2\RuinedFooocus\modules\async_worker.py", line 57, in handler
process(gen_data)
File "F:\RuinedFooocus_win64_1-25-2\RuinedFooocus\modules\async_worker.py", line 240, in process
imgs = pipeline.process(
File "f:\RuinedFooocus_win64_1-25-2\python_embeded\lib\site-packages\torch\utils_contextlib.py", line 115, in decorate_context
return func(*args, **kwargs)
File "F:\RuinedFooocus_win64_1-25-2\RuinedFooocus\modules\sdxl_pipeline.py", line 280, in process
if self.textencode("+", positive_prompt, clip_skip):
File "F:\RuinedFooocus_win64_1-25-2\RuinedFooocus\modules\sdxl_pipeline.py", line 202, in textencode
self.conditions[id]["cache"] = CLIPTextEncode().encode(
File "F:\RuinedFooocus_win64_1-25-2\RuinedFooocus\repositories\ComfyUI-from-StabilityAI-Official\nodes.py", line 58, in encode
cond, pooled = clip.encode_from_tokens(tokens, return_pooled=True)
File "F:\RuinedFooocus_win64_1-25-2\RuinedFooocus\repositories\ComfyUI-from-StabilityAI-Official\comfy\sd.py", line 142, in encode_from_tokens
cond, pooled = self.cond_stage_model.encode_token_weights(tokens)
File "F:\RuinedFooocus_win64_1-25-2\RuinedFooocus\repositories\ComfyUI-from-StabilityAI-Official\comfy\sd3_clip.py", line 124, in encode_token_weights
t5_out, t5_pooled = self.t5xxl.encode_token_weights(token_weight_pars_t5)
File "F:\RuinedFooocus_win64_1-25-2\RuinedFooocus\repositories\ComfyUI-from-StabilityAI-Official\comfy\sd1_clip.py", line 40, in encode_token_weights
out, pooled = self.encode(to_encode)
File "F:\RuinedFooocus_win64_1-25-2\RuinedFooocus\repositories\ComfyUI-from-StabilityAI-Official\comfy\sd1_clip.py", line 201, in encode
return self(tokens)
File "f:\RuinedFooocus_win64_1-25-2\python_embeded\lib\site-packages\torch\nn\modules\module.py", line 1501, in _call_impl
return forward_call(*args, **kwargs)
File "F:\RuinedFooocus_win64_1-25-2\RuinedFooocus\repositories\ComfyUI-from-StabilityAI-Official\comfy\sd1_clip.py", line 186, in forward
z = outputs[1].float()
AttributeError: 'NoneType' object has no attribute 'float'
Any news?
yeah.. doesnt work :D