Problems running A1111 on Google Collab

[williamsblood]

Even after Google banned SD from Colab, I was able to use it until yesterday (Via Colab Pro) but today when I tried to run the notebook I got this error:

WARNING[XFORMERS]: xFormers can't load C++/CUDA extensions. xFormers was built for:
PyTorch 2.0.1+cu118 with CUDA 1108 (you have 2.1.0+cu118)
Python 3.10.11 (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

Could anyone help?

[kiske-]

I got the same error.
Should I downgrade Python? But I don't know how to do that...

[Marsh-6]

Getting the same error since today. Did some research and seems like the solution is simple, but I just don't have enough code knowledge to implement it. Tried a clean reinstall but got the same error.

If someone could let us know how to implement this solution: facebookresearch/xformers#664

Thank you! :)

[williamsblood]

A working around here: #2615 (comment)

[kiske-]

worked for me! thanks!

[Marsh-6]

Worked yesterday but stopped working again today :(

[TheLastBen]

I hope his doggo (from his avatar) is fine.

She left this world a few years ago, sad world after that.

[ClaraSanders]

I'm really sorry. I know very well pets are irreplaceable but I hope another one fins its place in your heart.

[TheLastBen]

Use the latest notebook

[toeythewolf04]

Thank you for your updated sir. It's better but still error occur.

[toeythewolf04]

Tried to run and it's mostly done. But in the end of process. error occur after proceed. This was my error message >>

NotImplementedError: No operator found for memory_efficient_attention_forward with inputs: query : shape=(1, 5525, 1, 512) (torch.float32) key : shape=(1, 5525, 1, 512) (torch.float32) value : shape=(1, 5525, 1, 512) (torch.float32) attn_bias : <class 'NoneType'> p : 0.0 decoderF is not supported because: max(query.shape[-1] != value.shape[-1]) > 128 xFormers wasn't build with CUDA support attn_bias type is <class 'NoneType'> operator wasn't built - see python -m xformers.info for more info [email protected] is not supported because: max(query.shape[-1] != value.shape[-1]) > 256 xFormers wasn't build with CUDA support requires device with capability > (8, 0) but your GPU has capability (7, 5) (too old) dtype=torch.float32 (supported: {torch.bfloat16, torch.float16}) operator wasn't built - see python -m xformers.info for more info tritonflashattF is not supported because: max(query.shape[-1] != value.shape[-1]) > 128 xFormers wasn't build with CUDA support requires device with capability > (8, 0) but your GPU has capability (7, 5) (too old) dtype=torch.float32 (supported: {torch.bfloat16, torch.float16}) operator wasn't built - see python -m xformers.info for more info triton is not available requires GPU with sm80 minimum compute capacity, e.g., A100/H100/L4 cutlassF is not supported because: xFormers wasn't build with CUDA support operator wasn't built - see python -m xformers.info for more info smallkF is not supported because: max(query.shape[-1] != value.shape[-1]) > 32 xFormers wasn't build with CUDA support operator wasn't built - see python -m xformers.info for more info unsupported embed per head: 512

[toeythewolf04]

OH, I saw the updated.

no module 'xformers'. Processing without...

I can run it now.
Thank you very much sir.

[TheLastBen]

Use the latest notebook, all is fixed now