| Name | bitsandbytes JSON |
| Version |
0.45.1
JSON |
| download |
| home_page | None |
| Summary | k-bit optimizers and matrix multiplication routines. |
| upload_time | 2025-01-23 16:34:33 |
| maintainer | None |
| docs_url | None |
| author | None |
| requires_python | >=3.8 |
| license | MIT License Copyright (c) Facebook, Inc. and its affiliates. Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. |
| keywords |
gpu
optimizers
optimization
8-bit
quantization
compression
|
| VCS |
 |
| bugtrack_url |
|
| requirements |
No requirements were recorded.
|
| Travis-CI |
No Travis.
|
| coveralls test coverage |
No coveralls.
|
# `bitsandbytes`
[](https://pepy.tech/project/bitsandbytes) [](https://pepy.tech/project/bitsandbytes) [](https://pepy.tech/project/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](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:
- 🔥 **AMD GPUs** (ROCm)
- ⚡ **Intel CPUs** & **GPUs**
We’d love your early feedback! 🙏
👉 [Instructions for your `pip install` here](https://huggingface.co/docs/bitsandbytes/main/en/installation#multi-backend)
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 🤗
## License
`bitsandbytes` is MIT licensed.
We thank Fabio Cannizzo for his work on [FastBinarySearch](https://github.com/fabiocannizzo/FastBinarySearch) which we use for CPU quantization.
Raw data
{
"_id": null,
"home_page": null,
"name": "bitsandbytes",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.8",
"maintainer_email": "Titus von K\u00f6ller <titus@huggingface.co>, Matthew Douglas <matthew.douglas@huggingface.co>",
"keywords": "gpu, optimizers, optimization, 8-bit, quantization, compression",
"author": null,
"author_email": "Tim Dettmers <dettmers@cs.washington.edu>",
"download_url": null,
"platform": null,
"description": "# `bitsandbytes`\n\n[](https://pepy.tech/project/bitsandbytes) [](https://pepy.tech/project/bitsandbytes) [](https://pepy.tech/project/bitsandbytes)\n\nThe `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.\n\nThe 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.\n\nThere 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.\n\n**Please head to the official documentation page:**\n\n**[https://huggingface.co/docs/bitsandbytes/main](https://huggingface.co/docs/bitsandbytes/main)**\n\n## `bitsandbytes` multi-backend _alpha_ release is out!\n\n\ud83d\ude80 Big news! After months of hard work and incredible community contributions, we're thrilled to announce the **bitsandbytes multi-backend _alpha_ release**! \ud83d\udca5\n\nNow supporting:\n- \ud83d\udd25 **AMD GPUs** (ROCm)\n- \u26a1 **Intel CPUs** & **GPUs**\n\nWe\u2019d love your early feedback! \ud83d\ude4f\n\n\ud83d\udc49 [Instructions for your `pip install` here](https://huggingface.co/docs/bitsandbytes/main/en/installation#multi-backend)\n\nWe'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 \ud83e\udd17\n\n## License\n\n`bitsandbytes` is MIT licensed.\n\nWe thank Fabio Cannizzo for his work on [FastBinarySearch](https://github.com/fabiocannizzo/FastBinarySearch) which we use for CPU quantization.\n",
"bugtrack_url": null,
"license": "MIT License Copyright (c) Facebook, Inc. and its affiliates. Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the \"Software\"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED \"AS IS\", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. ",
"summary": "k-bit optimizers and matrix multiplication routines.",
"version": "0.45.1",
"project_urls": {
"changelog": "https://github.com/bitsandbytes-foundation/bitsandbytes/blob/main/CHANGELOG.md",
"docs": "https://huggingface.co/docs/bitsandbytes/main",
"homepage": "https://github.com/bitsandbytes-foundation/bitsandbytes",
"issues": "https://github.com/bitsandbytes-foundation/bitsandbytes/issues"
},
"split_keywords": [
"gpu",
" optimizers",
" optimization",
" 8-bit",
" quantization",
" compression"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "e4260897c35865a6d7a04b05af5d59558484fda14fa26e5ec922a3bb8d3e0d18",
"md5": "724bdd02128d7c5e443146df990e8552",
"sha256": "f9f6a38b3fccd012549f875e4f371a47c515c4394df7a488dec65faa14465323"
},
"downloads": -1,
"filename": "bitsandbytes-0.45.1-py3-none-manylinux_2_24_x86_64.whl",
"has_sig": false,
"md5_digest": "724bdd02128d7c5e443146df990e8552",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.8",
"size": 69679919,
"upload_time": "2025-01-23T16:34:33",
"upload_time_iso_8601": "2025-01-23T16:34:33.810177Z",
"url": "https://files.pythonhosted.org/packages/e4/26/0897c35865a6d7a04b05af5d59558484fda14fa26e5ec922a3bb8d3e0d18/bitsandbytes-0.45.1-py3-none-manylinux_2_24_x86_64.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "f136022554463c267f2e3152a17a8bbd979cbf5ae19163bc6958796ddd2d6443",
"md5": "ed6daa040969826965b9b929d0a0f042",
"sha256": "3f5b8c982ae475a7fee2f21c2b1e7835d212da0db10bb5e270a3508614ee2cf9"
},
"downloads": -1,
"filename": "bitsandbytes-0.45.1-py3-none-win_amd64.whl",
"has_sig": false,
"md5_digest": "ed6daa040969826965b9b929d0a0f042",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.8",
"size": 69124162,
"upload_time": "2025-01-23T16:34:44",
"upload_time_iso_8601": "2025-01-23T16:34:44.921858Z",
"url": "https://files.pythonhosted.org/packages/f1/36/022554463c267f2e3152a17a8bbd979cbf5ae19163bc6958796ddd2d6443/bitsandbytes-0.45.1-py3-none-win_amd64.whl",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2025-01-23 16:34:33",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "bitsandbytes-foundation",
"github_project": "bitsandbytes",
"travis_ci": false,
"coveralls": false,
"github_actions": true,
"lcname": "bitsandbytes"
}
