[FIXED] NotImplementedError: No operator found for memory_efficient_attention_forward with inputs

[rantianhua]

I'm a beginner to try unsloth. I run the free notebook Llama 3 (8B), and then got the following error:

I also encountered the following error during the first installing step:

ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.
xformers 0.0.26.post1 requires torch==2.3.0, but you have torch 2.2.1+cu121 which is incompatible.

[tariksghiouri]

facing the same issue , I have no idea why.
the cell was running just a couple of days ago.

[jamiehughes5926]

facing the same issue

[danielhanchen]

@rantianhua @tariksghiouri @jamiehughes5926 Apologies on the issue! I should have wrote it here - please update the first cell's installatation instructions from

%%capture
import torch
major_version, minor_version = torch.cuda.get_device_capability()
# Must install separately since Colab has torch 2.2.1, which breaks packages
!pip install "unsloth[colab-new] @ git+https://github.com/unslothai/unsloth.git"
if major_version >= 8:
    # Use this for new GPUs like Ampere, Hopper GPUs (RTX 30xx, RTX 40xx, A100, H100, L40)
    !pip install --no-deps packaging ninja einops flash-attn xformers trl peft accelerate bitsandbytes
else:
    # Use this for older GPUs (V100, Tesla T4, RTX 20xx)
    !pip install --no-deps xformers trl peft accelerate bitsandbytes
pass

to

%%capture
# Installs Unsloth, Xformers (Flash Attention) and all other packages!
!pip install "unsloth[colab-new] @ git+https://github.com/unslothai/unsloth.git"
!pip install --no-deps "xformers<0.0.26" trl peft accelerate bitsandbytes

[tariksghiouri]

thank you Daniel :)