# AutoAWQ Kernels
AutoAWQ Kernels is a new package that is split up from the [main repository](https://github.com/casper-hansen/AutoAWQ) in order to avoid compilation times.
## Requirements
- Windows: Must use WSL2.
- NVIDIA:
- GPU: Must be compute capability 7.5 or higher.
- CUDA Toolkit: Must be 11.8 or higher.
- AMD:
- ROCm: Must be 5.6 or higher. [Build from source](#build-from-source)
## Install
### Install from PyPi
The package is available on PyPi with CUDA 12.4.1 wheels:
```
pip install autoawq-kernels
```
### Build from source
To build the kernels from source, you first need to setup an environment containing the necessary dependencies.
#### Build Requirements
- Python>=3.8.0
- Numpy
- Wheel
- PyTorch
- ROCm: You need to install the following packages `rocsparse-dev hipsparse-dev rocthrust-dev rocblas-dev hipblas-dev`.
#### Building process
```
pip install git+https://github.com/casper-hansen/AutoAWQ_kernels.git
```
Notes on environment variables:
- `TORCH_VERSION`: By default, we build using the current version of torch by `torch.__version__`. You can override it with `TORCH_VERSION`.
- `CUDA_VERSION` or `ROCM_VERSION` can also be used to build for a specific version of CUDA or ROCm.
- `CC` and `CXX`: You can specify which build system to use for the C code, e.g. `CC=g++-13 CXX=g++-13 pip install -e .`
- `COMPUTE_CAPABILITIES`: You can specify specific compute capabilities to compile for: `COMPUTE_CAPABILITIES="75,80,86,87,89,90" pip install -e .`
Raw data
{
"_id": null,
"home_page": "https://github.com/casper-hansen/AutoAWQ_kernels",
"name": "autoawq-kernels",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.8.0",
"maintainer_email": null,
"keywords": "awq, autoawq, quantization, transformers",
"author": "Casper Hansen",
"author_email": null,
"download_url": null,
"platform": "linux",
"description": "# AutoAWQ Kernels\n\nAutoAWQ Kernels is a new package that is split up from the [main repository](https://github.com/casper-hansen/AutoAWQ) in order to avoid compilation times.\n\n## Requirements\n\n- Windows: Must use WSL2.\n\n- NVIDIA:\n - GPU: Must be compute capability 7.5 or higher.\n - CUDA Toolkit: Must be 11.8 or higher.\n- AMD:\n - ROCm: Must be 5.6 or higher. [Build from source](#build-from-source)\n\n## Install\n\n### Install from PyPi\n\nThe package is available on PyPi with CUDA 12.4.1 wheels:\n\n```\npip install autoawq-kernels\n```\n\n### Build from source\n\nTo build the kernels from source, you first need to setup an environment containing the necessary dependencies.\n\n#### Build Requirements\n\n- Python>=3.8.0\n- Numpy\n- Wheel\n- PyTorch\n- ROCm: You need to install the following packages `rocsparse-dev hipsparse-dev rocthrust-dev rocblas-dev hipblas-dev`.\n\n#### Building process\n\n```\npip install git+https://github.com/casper-hansen/AutoAWQ_kernels.git\n```\n\nNotes on environment variables:\n- `TORCH_VERSION`: By default, we build using the current version of torch by `torch.__version__`. You can override it with `TORCH_VERSION`.\n - `CUDA_VERSION` or `ROCM_VERSION` can also be used to build for a specific version of CUDA or ROCm.\n- `CC` and `CXX`: You can specify which build system to use for the C code, e.g. `CC=g++-13 CXX=g++-13 pip install -e .`\n- `COMPUTE_CAPABILITIES`: You can specify specific compute capabilities to compile for: `COMPUTE_CAPABILITIES=\"75,80,86,87,89,90\" pip install -e .`\n",
"bugtrack_url": null,
"license": "MIT",
"summary": "AutoAWQ Kernels implements the AWQ kernels.",
"version": "0.0.9",
"project_urls": {
"Homepage": "https://github.com/casper-hansen/AutoAWQ_kernels"
},
"split_keywords": [
"awq",
" autoawq",
" quantization",
" transformers"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "98a6c48cf823c2d29731ae262a05e17d317165df7bef68a486e5840405b70cc0",
"md5": "667f6d43ceb618a0a8f0a77f07ad8b83",
"sha256": "ed8f4d744df21beae445efb1de54061bffc5fccbfefc8ae65c1dc10d08f90052"
},
"downloads": -1,
"filename": "autoawq_kernels-0.0.9-cp310-cp310-manylinux2014_x86_64.whl",
"has_sig": false,
"md5_digest": "667f6d43ceb618a0a8f0a77f07ad8b83",
"packagetype": "bdist_wheel",
"python_version": "cp310",
"requires_python": ">=3.8.0",
"size": 37352547,
"upload_time": "2024-11-16T15:42:59",
"upload_time_iso_8601": "2024-11-16T15:42:59.746190Z",
"url": "https://files.pythonhosted.org/packages/98/a6/c48cf823c2d29731ae262a05e17d317165df7bef68a486e5840405b70cc0/autoawq_kernels-0.0.9-cp310-cp310-manylinux2014_x86_64.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "2e35471f80543e31aef5097b037cc9626a47a0edb20311c3951060673cc83868",
"md5": "c5aa60c9ea4f650e21f75f532dfdf619",
"sha256": "cd7d3db501068b3a12094a07921d985a57e640725cdda1252d4b135ed6aeaa65"
},
"downloads": -1,
"filename": "autoawq_kernels-0.0.9-cp310-cp310-win_amd64.whl",
"has_sig": false,
"md5_digest": "c5aa60c9ea4f650e21f75f532dfdf619",
"packagetype": "bdist_wheel",
"python_version": "cp310",
"requires_python": ">=3.8.0",
"size": 2747752,
"upload_time": "2024-11-16T15:43:03",
"upload_time_iso_8601": "2024-11-16T15:43:03.134798Z",
"url": "https://files.pythonhosted.org/packages/2e/35/471f80543e31aef5097b037cc9626a47a0edb20311c3951060673cc83868/autoawq_kernels-0.0.9-cp310-cp310-win_amd64.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "4eb0e9f7142e58e5539892cb0558e9a24d894f095a9904b4014f892a43c39229",
"md5": "896e8f8891837ddc5681ec1e6e631034",
"sha256": "fe800a6538691afaa77abe7c8b2b0a121351843f048d54e11d617d604dcba48f"
},
"downloads": -1,
"filename": "autoawq_kernels-0.0.9-cp311-cp311-manylinux2014_x86_64.whl",
"has_sig": false,
"md5_digest": "896e8f8891837ddc5681ec1e6e631034",
"packagetype": "bdist_wheel",
"python_version": "cp311",
"requires_python": ">=3.8.0",
"size": 37377774,
"upload_time": "2024-11-16T15:43:08",
"upload_time_iso_8601": "2024-11-16T15:43:08.917469Z",
"url": "https://files.pythonhosted.org/packages/4e/b0/e9f7142e58e5539892cb0558e9a24d894f095a9904b4014f892a43c39229/autoawq_kernels-0.0.9-cp311-cp311-manylinux2014_x86_64.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "8658093d5cd5d48f82aeb29ab0dcf76c3ed1a636c2c249ab517773190cb77a67",
"md5": "c7d056e76910506db6113eb93875dbed",
"sha256": "8c7f2404b3aa448ff77872dd6ba2963ce8b685d8aa73ef65fd1b8bc85d92b17d"
},
"downloads": -1,
"filename": "autoawq_kernels-0.0.9-cp311-cp311-win_amd64.whl",
"has_sig": false,
"md5_digest": "c7d056e76910506db6113eb93875dbed",
"packagetype": "bdist_wheel",
"python_version": "cp311",
"requires_python": ">=3.8.0",
"size": 2753717,
"upload_time": "2024-11-16T15:43:12",
"upload_time_iso_8601": "2024-11-16T15:43:12.826500Z",
"url": "https://files.pythonhosted.org/packages/86/58/093d5cd5d48f82aeb29ab0dcf76c3ed1a636c2c249ab517773190cb77a67/autoawq_kernels-0.0.9-cp311-cp311-win_amd64.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "4aef613310287e009d35b740b76b96913ea87e5522d7f91845bf817d4ed0abd9",
"md5": "f5252fdc2802b899d3f13dfdf7634faf",
"sha256": "4c41a71af1d5a75e52c9833b9c48237b04d3b0eee26d712fc1b074af9135afc8"
},
"downloads": -1,
"filename": "autoawq_kernels-0.0.9-cp312-cp312-manylinux2014_x86_64.whl",
"has_sig": false,
"md5_digest": "f5252fdc2802b899d3f13dfdf7634faf",
"packagetype": "bdist_wheel",
"python_version": "cp312",
"requires_python": ">=3.8.0",
"size": 37423485,
"upload_time": "2024-11-16T15:43:18",
"upload_time_iso_8601": "2024-11-16T15:43:18.560890Z",
"url": "https://files.pythonhosted.org/packages/4a/ef/613310287e009d35b740b76b96913ea87e5522d7f91845bf817d4ed0abd9/autoawq_kernels-0.0.9-cp312-cp312-manylinux2014_x86_64.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "dca32a1966f685a980c1ad5662f92a1aa1a84608fb376b534301330487fc1db8",
"md5": "9837d7e837a0349592b8ecac37c8f674",
"sha256": "f259e7c60b11fa0689bb337dd4456319787256cbd2a8e4a491f01b51bb6c43d1"
},
"downloads": -1,
"filename": "autoawq_kernels-0.0.9-cp312-cp312-win_amd64.whl",
"has_sig": false,
"md5_digest": "9837d7e837a0349592b8ecac37c8f674",
"packagetype": "bdist_wheel",
"python_version": "cp312",
"requires_python": ">=3.8.0",
"size": 2755850,
"upload_time": "2024-11-16T15:43:22",
"upload_time_iso_8601": "2024-11-16T15:43:22.767087Z",
"url": "https://files.pythonhosted.org/packages/dc/a3/2a1966f685a980c1ad5662f92a1aa1a84608fb376b534301330487fc1db8/autoawq_kernels-0.0.9-cp312-cp312-win_amd64.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "45bb4554a1b3cf0d29acf8ef449e875ae24c6935d68858876b0ec44a2fd4c2e5",
"md5": "759cd2a1d37a14f2c999b34481fbd2ff",
"sha256": "b6baf039c22deb02f2ae194fdd77551b3c85c8f8a77b749f7caa17dacf986adb"
},
"downloads": -1,
"filename": "autoawq_kernels-0.0.9-cp39-cp39-manylinux2014_x86_64.whl",
"has_sig": false,
"md5_digest": "759cd2a1d37a14f2c999b34481fbd2ff",
"packagetype": "bdist_wheel",
"python_version": "cp39",
"requires_python": ">=3.8.0",
"size": 37336723,
"upload_time": "2024-11-16T15:43:29",
"upload_time_iso_8601": "2024-11-16T15:43:29.135671Z",
"url": "https://files.pythonhosted.org/packages/45/bb/4554a1b3cf0d29acf8ef449e875ae24c6935d68858876b0ec44a2fd4c2e5/autoawq_kernels-0.0.9-cp39-cp39-manylinux2014_x86_64.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "3b6062b1ff1406dd4fdd90ad8dce840f8e4c19f01b755cad86d6498e55dfa373",
"md5": "f45a15e6ebd21cd136d2857e3e5115f8",
"sha256": "6ad12dd68b0932182678f2f9fbee87452707b81f0e8dad242d23af018358f030"
},
"downloads": -1,
"filename": "autoawq_kernels-0.0.9-cp39-cp39-win_amd64.whl",
"has_sig": false,
"md5_digest": "f45a15e6ebd21cd136d2857e3e5115f8",
"packagetype": "bdist_wheel",
"python_version": "cp39",
"requires_python": ">=3.8.0",
"size": 2746275,
"upload_time": "2024-11-16T15:43:32",
"upload_time_iso_8601": "2024-11-16T15:43:32.324206Z",
"url": "https://files.pythonhosted.org/packages/3b/60/62b1ff1406dd4fdd90ad8dce840f8e4c19f01b755cad86d6498e55dfa373/autoawq_kernels-0.0.9-cp39-cp39-win_amd64.whl",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-11-16 15:42:59",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "casper-hansen",
"github_project": "AutoAWQ_kernels",
"travis_ci": false,
"coveralls": false,
"github_actions": true,
"lcname": "autoawq-kernels"
}
JSON
