bitsandbytes 0.43.1

k-bit optimizers and matrix multiplication routines.

Stars: 5526, Watchers: 5526, Forks: 570, Open Issues: 145

The TimDettmers/bitsandbytes repo was created 2 years ago and the last code push was 12 hours ago.
The project is extremely popular with a mindblowing 5526 github stars!

How to Install bitsandbytes

You can install bitsandbytes using pip

pip install bitsandbytes

or add it to a project with poetry

poetry add bitsandbytes

Package Details

Author

Tim Dettmers

License

MIT

Homepage

https://github.com/TimDettmers/bitsandbytes

PyPi:

https://pypi.org/project/bitsandbytes/

GitHub Repo:

https://github.com/TimDettmers/bitsandbytes

Classifiers

• Scientific/Engineering/Artificial Intelligence

Related Packages

Errors

A list of common bitsandbytes errors.

Code Examples

Here are some bitsandbytes code examples and snippets.

GitHub Issues

The bitsandbytes package has 145 open issues on GitHub

• Getting errors when attempting to run on remote cluster
• UserWarning: WARNING: Compute capability < 7.5 detected! Only slow 8-bit matmul is supported for your GPU! warn(msg) Detected 8-bit loading: activating 8-bit loading for this model
• RuntimeError: Something when wrong when trying to find file. Maybe you do not have a linux system?
• add sort to get_compute_capabilities
• Update cuda_install.sh
• Added scipy to requirements.txt
• error: identifier "__cudaDeviceSynchronizeDeprecationAvoidance" is undefined
• Need help with train error after install
• AdamW8 bit and Lion8bit not working - CUDA_SETUP: WARNING! libcudart.so not found in any environmental path.
• Cuda ERROR
• Example throws Tokenizer class LLaMATokenizer does not exist
• 3.5x slow 4bit inference for tiiuae/falcon-7b-instruct on RTX 3060 as compared to 8bit
• CUDA exception! Error code: no CUDA-capable device is detected
• AttributeError: module 'bitsandbytes.nn' has no attribute 'Linear8bitLt'
• what's the differences between torch.optim.Adam and bnb.nn.Adam32bit?

See more issues on GitHub