ModuleNotFoundError: No module named 'triton'

[boseong-yun]

I have been using the codes for a while (at least a week ago), and I can no longer import the module. There seem to be other people experiencing the same issues, but was not sure whether this problem was exactly the same. Sorry if this is a repeat. Here are the details:

After running the following (directly copied from your example code):

%%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<0.0.27" "trl<0.9.0" peft accelerate bitsandbytes

The following code chunk throws the error

Code:

from unsloth import FastLanguageModel

Errror:

---------------------------------------------------------------------------
ModuleNotFoundError                       Traceback (most recent call last)
[<ipython-input-15-ff6398fc40c7>](https://localhost:8080/#) in <cell line: 1>()
----> 1 from unsloth import FastLanguageModel

[/usr/local/lib/python3.10/dist-packages/unsloth/__init__.py](https://localhost:8080/#) in <module>
     97 import bitsandbytes as bnb
     98 
---> 99 import triton
    100 libcuda_dirs = lambda: None
    101 if Version(triton.__version__) >= Version("3.0.0"):

ModuleNotFoundError: No module named 'triton'

---------------------------------------------------------------------------

[danielhanchen]

Apologies on the issue!
Colab just updated to torch 2.4 - please change the top installation cell to

%%capture
# Installs Unsloth, Xformers (Flash Attention) and all other packages!
!pip install "unsloth[colab-new] @ git+https://github.com/unslothai/unsloth.git"

# We have to check which Torch version for Xformers (2.3 -> 0.0.27)
from torch import __version__; from packaging.version import Version as V
xformers = "xformers==0.0.27" if V(__version__) < V("2.4.0") else "xformers"
!pip install --no-deps {xformers} trl peft accelerate bitsandbytes triton

[boseong-yun]

Works fine! Thank you for your prompt help.

[patruff]

Had same issue thanks for the fix

[kebbbnnn]

Thanks for this! Though it spewed a different error when I first applied the fix in my colab notebook, deleting the runtime and connecting to it again fixed it.

[danielhanchen]

No problems and sorry again!