bitsandbytes
The bitsandbytes library is a lightweight Python wrapper around CUDA custom functions, in particular 8-bit optimizers, matrix multiplication (LLM.int8()), and 8 & 4-bit quantization functions.
The library includes quantization primitives for 8-bit & 4-bit operations, through bitsandbytes.nn.Linear8bitLt and bitsandbytes.nn.Linear4bit and 8-bit optimizers through bitsandbytes.optim module.
There are ongoing efforts to support further hardware backends, i.e. Intel CPU + GPU, AMD GPU, Apple Silicon. Windows support is quite far along and is on its way as well.
Please head to the official documentation page:
https://huggingface.co/docs/bitsandbytes/main
bitsandbytes multi-backend alpha release is out!🚀 Big news! After months of hard work and incredible community contributions, we're thrilled to announce the bitsandbytes multi-backend alpha release! 💥
Now supporting:
We’d love your early feedback! 🙏
👉 Instructions for your pip install here
We're super excited about these recent developments and grateful for any constructive input or support that you can give to help us make this a reality (e.g. helping us with the upcoming Apple Silicon backend or reporting bugs). BNB is a community project and we're excited for your collaboration 🤗
bitsandbytes is MIT licensed.
We thank Fabio Cannizzo for his work on FastBinarySearch which we use for CPU quantization.
FAQs
k-bit optimizers and matrix multiplication routines.
We found that bitsandbytes demonstrated a healthy version release cadence and project activity because the last version was released less than a year ago. It has 2 open source maintainers collaborating on the project.
Did you know?
Socket for GitHub automatically highlights issues in each pull request and monitors the health of all your open source dependencies. Discover the contents of your packages and block harmful activity before you install or update your dependencies.
Security News
pnpm 10 blocks lifecycle scripts by default to improve security, addressing supply chain attack risks but sparking debate over compatibility and workflow changes.
Product
Socket now supports uv.lock files to ensure consistent, secure dependency resolution for Python projects and enhance supply chain security.
Research
Security News
Socket researchers have discovered multiple malicious npm packages targeting Solana private keys, abusing Gmail to exfiltrate the data and drain Solana wallets.