Home Projects bitsandbytes
bitsandbytes
Python

bitsandbytes

Accessible large language models via k-bit quantization for PyTorch.

by bitsandbytes-foundation · GitHub
Stars
Forks
License
Created
Last commit
Category
Language
llmmachine-learningpytorchMITPython
View on GitHub
In plain words

Use less memory to run and train large language models in PyTorch without losing performance.

You maintain this project?

Claim its page: indexed whatever its rank, translated into six languages, and enriched with what you write yourself.

Claim this page →
bitsandbytes — GitHub preview card
📈 Star history
8.4k8.3k
2026-07-042026-08-31
📈 Track bitsandbytes

Get an email alert on its next release or when it starts trending — never miss the moment.

Free · no card · unsubscribe anytime
Get email alerts →
📄 About

Accessible large language models via k-bit quantization for PyTorch.

bitsandbytes has 8.4k stars on GitHub. It has been forked 919 times. bitsandbytes is written mainly in Python. It has been in active development since 2021. bitsandbytes is available under the MIT license. Its main topics are llm, machine-learning, pytorch, qlora.

Frequently asked questions

What is bitsandbytes?

Accessible large language models via k-bit quantization for PyTorch.

Is bitsandbytes open source?

bitsandbytes is an open-source project. It is released under the MIT license.

Is bitsandbytes free?

Yes. bitsandbytes is free and open source — you can use, modify and self-host it.

What license does bitsandbytes use?

bitsandbytes is available under the MIT license.

What language is bitsandbytes written in?

bitsandbytes is written mainly in Python.

🏅 Maintainer of this project?
olud.ai badge — bitsandbytes

Add this live badge to your README — your GitHub stars and directory rank, refreshed daily.

[![olud.ai](https://olud.ai/badge.php?tool=bitsandbytes-foundation-bitsandbytes)](https://olud.ai/project/bitsandbytes-foundation-bitsandbytes.html)
More badge options →
🧬 Shares DNA with🧬 View the DNA map →

Measured from GitHub topics shared by both projects, weighted by how rare each topic is.