• Repository

Meta

• License Expression: MIT
  SPDX License Expression
• Author: Tim Dettmers
• Maintainer: Titus von Köller
• Tags gpu , optimizers , optimization , 8-bit , quantization , compression
• Requires: Python >=3.10
• Provides-Extra: benchmark , docs , dev , test

Classifiers

• Development Status
  • 4 - Beta
• Environment
  • GPU :: NVIDIA CUDA :: 11.8
  • GPU :: NVIDIA CUDA :: 12
  • GPU :: NVIDIA CUDA :: 13
• Intended Audience
  • Developers
  • Science/Research
• Operating System
  • MacOS
  • Microsoft :: Windows
  • POSIX :: Linux
• Programming Language
  • C++
  • Python :: 3.10
  • Python :: 3.11
  • Python :: 3.12
  • Python :: 3.13
  • Python :: 3.14
  • Python :: Implementation :: CPython
• Topic
  • Scientific/Engineering :: Artificial Intelligence

👁 Image

bitsandbytes

👁 License
👁 Downloads
👁 Nightly Unit Tests
👁 GitHub Release
👁 PyPI - Python Version

bitsandbytes enables accessible large language models via k-bit quantization for PyTorch. We provide three main features for dramatically reducing memory consumption for inference and training:

• 8-bit optimizers uses block-wise quantization to maintain 32-bit performance at a small fraction of the memory cost.
• LLM.int8() or 8-bit quantization enables large language model inference with only half the required memory and without any performance degradation. This method is based on vector-wise quantization to quantize most features to 8-bits and separately treating outliers with 16-bit matrix multiplication.
• QLoRA or 4-bit quantization enables large language model training with several memory-saving techniques that don't compromise performance. This method quantizes a model to 4-bits and inserts a small set of trainable low-rank adaptation (LoRA) weights to allow training.

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.

System Requirements

bitsandbytes has the following minimum requirements for all platforms:

• Python 3.10+
• PyTorch 2.3+
  • Note: While we aim to provide wide backwards compatibility, we recommend using the latest version of PyTorch for the best experience.

Accelerator support:

Note: this table reflects the status of the current development branch. For the latest stable release, see the document in the 0.49.2 tag.

Legend:

🚧 = In Development, 〰️ = Partially Supported, ✅ = Supported, 🐢 = Slow Implementation Supported, ❌ = Not Supported

Platform	Accelerator	Hardware Requirements	LLM.int8()	QLoRA 4-bit	8-bit Optimizers
🐧 Linux, glibc >= 2.24
x86-64	◻️ CPU	Minimum: AVX2 Optimized: AVX512F, AVX512BF16	✅	✅	❌
🟩 NVIDIA GPU cuda	SM60+ minimum SM75+ recommended	✅	✅	✅
🟥 AMD GPU cuda	CDNA: gfx90a, gfx942, gfx950 RDNA: gfx1100, gfx1101, gfx1150, gfx1151, gfx1200, gfx1201	✅	✅	✅
🟦 Intel GPU xpu	Data Center GPU Max Series Arc A-Series (Alchemist) Arc B-Series (Battlemage)	✅	✅	〰️
🟪 Intel Gaudi hpu	Gaudi2, Gaudi3	✅	〰️	❌
aarch64	◻️ CPU	✅	✅	❌
🟩 NVIDIA GPU cuda	SM75+	✅	✅	✅
🪟 Windows 11 / Windows Server 2022+
x86-64	◻️ CPU	AVX2	✅	✅	❌
🟩 NVIDIA GPU cuda	SM60+ minimum SM75+ recommended	✅	✅	✅
🟦 Intel GPU xpu	Arc A-Series (Alchemist) Arc B-Series (Battlemage)	✅	✅	〰️
🍎 macOS 14+
arm64	◻️ CPU	Apple M1+	✅	✅	❌
⬜ Metal mps	Apple M1+	🐢	🐢	❌

:book: Documentation

• Official Documentation
• 🤗 Transformers
• 🤗 Diffusers
• 🤗 PEFT

:heart: Sponsors

The continued maintenance and development of bitsandbytes is made possible thanks to the generous support of our sponsors. Their contributions help ensure that we can keep improving the project and delivering valuable updates to the community.

License

bitsandbytes is MIT licensed.

How to cite us

If you found this library useful, please consider citing our work:

QLoRA

@article{dettmers2023qlora,
title={Qlora: Efficient finetuning of quantized llms},
author={Dettmers, Tim and Pagnoni, Artidoro and Holtzman, Ari and Zettlemoyer, Luke},
journal={arXiv preprint arXiv:2305.14314},
year={2023}
}

LLM.int8()

@article{dettmers2022llmint8,
title={LLM.int8(): 8-bit Matrix Multiplication for Transformers at Scale},
author={Dettmers, Tim and Lewis, Mike and Belkada, Younes and Zettlemoyer, Luke},
journal={arXiv preprint arXiv:2208.07339},
year={2022}
}

8-bit Optimizers

@article{dettmers2022optimizers,
title={8-bit Optimizers via Block-wise Quantization},
author={Dettmers, Tim and Lewis, Mike and Shleifer, Sam and Zettlemoyer, Luke},
journal={9th International Conference on Learning Representations, ICLR},
year={2022}
}

Verified details

These details have been verified by PyPI

Project links

• changelog
• homepage
• issues

GitHub Statistics

Maintainers

👁 Avatar for timdettmers from gravatar.com
timdettmers 👁 Avatar for Titus-von-Koeller from gravatar.com
Titus-von-Koeller

Unverified details

These details have not been verified by PyPI

Project links

• docs

• Download URL: bitsandbytes-0.49.2-py3-none-win_amd64.whl
• Upload date: Feb 16, 2026
• Size: 55.4 MB
• Tags: Python 3, Windows x86-64
• Uploaded using Trusted Publishing? Yes
• Uploaded via: twine/6.1.0 CPython/3.13.7

Hashes for bitsandbytes-0.49.2-py3-none-win_amd64.whl

Algorithm	Hash digest
SHA256	2e0ddd09cd778155388023cbe81f00afbb7c000c214caef3ce83386e7144df7d
MD5	24507e26015989c87580fe9e2763485c
BLAKE2b-256	b6d4501655842ad6771fb077f576d78cbedb5445d15b1c3c91343ed58ca46f0e

See more details on using hashes here.

• Download URL: bitsandbytes-0.49.2-py3-none-manylinux_2_24_x86_64.whl
• Upload date: Feb 16, 2026
• Size: 60.7 MB
• Tags: Python 3, manylinux: glibc 2.24+ x86-64
• Uploaded using Trusted Publishing? Yes
• Uploaded via: twine/6.1.0 CPython/3.13.7

Hashes for bitsandbytes-0.49.2-py3-none-manylinux_2_24_x86_64.whl

Algorithm	Hash digest
SHA256	54b771f06e1a3c73af5c7f16ccf0fc23a846052813d4b008d10cb6e017dd1c8c
MD5	3ffb359422a39b04b75835143a93409f
BLAKE2b-256	19573443d6f183436fbdaf5000aac332c4d5ddb056665d459244a5608e98ae92

See more details on using hashes here.

• Download URL: bitsandbytes-0.49.2-py3-none-manylinux_2_24_aarch64.whl
• Upload date: Feb 16, 2026
• Size: 31.4 MB
• Tags: Python 3, manylinux: glibc 2.24+ ARM64
• Uploaded using Trusted Publishing? Yes
• Uploaded via: twine/6.1.0 CPython/3.13.7

Hashes for bitsandbytes-0.49.2-py3-none-manylinux_2_24_aarch64.whl

Algorithm	Hash digest
SHA256	2fc0830c5f7169be36e60e11f2be067c8f812dfcb829801a8703735842450750
MD5	f9cf7b8ae1f5df4f9fdcab0b570d6ca9
BLAKE2b-256	2971acff7af06c818664aa87ff73e17a52c7788ad746b72aea09d3cb8e424348

See more details on using hashes here.

• Download URL: bitsandbytes-0.49.2-py3-none-macosx_14_0_arm64.whl
• Upload date: Feb 16, 2026
• Size: 131.9 kB
• Tags: Python 3, macOS 14.0+ ARM64
• Uploaded using Trusted Publishing? Yes
• Uploaded via: twine/6.1.0 CPython/3.13.7

Hashes for bitsandbytes-0.49.2-py3-none-macosx_14_0_arm64.whl

Algorithm	Hash digest
SHA256	87be5975edeac5396d699ecbc39dfc47cf2c026daaf2d5852a94368611a6823f
MD5	e205322e3ba50ecb82cc721917137b4b
BLAKE2b-256	d87df1fe0992334b18cd8494f89aeec1dcc674635584fcd9f115784fea3a1d05

See more details on using hashes here.