GitLab

* CUDA 13 build enablement

* Try to fix Windows build workflow

* Add torch 2.9+cu130 to tests

* Fix python version

* Update test workflow

* Don't test CPU on torch 2.9 yet

* Update doc

* Add CUDA 12.9 to build/test workflows

* Downgrade Jimver/cuda-toolkit to v0.2.24

* Update python-package.yml

* Update python-package.yml

* Update python-package.yml

* Update tests.yml

* Update tests.yml

* Add aarch64 cpu tests and CUDA build to nightly workflow

* aarch64: limit CUDA targets to sm75, sm80, sm90, sm100

* aarch64: limit CUDA targets to sm75, sm80, sm90, sm100

* Update build cpu script

* fix

* Update auditwheel for aarch64

* Run unit tests on GH Actions

* fix

* fix

* trigger workflow

* Update

* Update

* Update

* Run tests nightly

* Disable paged optimizer test on Windows

* Skip unit tests on Windows for CUDA 12.x (driver on runner is too old)

* Stop building for CUDA toolkit < 11.8

* Simplify

* Drop sm70 from cu128 build targets to align with pytorch

* initial support blackwell

* Update CHANGELOG.md
Co-authored-by: Matthew Douglas <38992547+matthewdouglas@users.noreply.github.com>

* Update CMakeLists.txt

* Update CHANGELOG.md

* fix build-cuda.sh

* fix build-cuda.sh

* fix cuda 12.7 build-cuda.sh

* Update build-cuda.sh

* Update cuda from 12.6.2 to 12.6.3

* Update .github/workflows/python-package.yml
Co-authored-by: Matthew Douglas <38992547+matthewdouglas@users.noreply.github.com>

* Update install_cuda.py
Co-authored-by: Matthew Douglas <38992547+matthewdouglas@users.noreply.github.com>

* Update install_cuda.sh
Co-authored-by: Matthew Douglas <38992547+matthewdouglas@users.noreply.github.com>

* Update .github/scripts/build-cuda.sh

* Update install_cuda.sh

---------
Co-authored-by: Matthew Douglas <38992547+matthewdouglas@users.noreply.github.com>

* Start of int8 refactor: remove col32/col_ampere/col_turing transforms in new igemmlt implementation

* Fix unintended change

* New naive mm_dequant kernel for row-major; cleanup

* fix

* int8 refactor: initial sparse decomp, cleanup

* Int8 refactoring: remove separate NO_CUBLASLT build; more cleanup

* int8: inference optimizations, some cleanup

* int8: more tests passing, cleanup

* int8 - more cleanup, most tests passing

* int8: specify CUDA stream for int8 ops

* perf: reduce overhead from getting cudaStream ptr

* Mark some functions for deprecation.

* int8 sparse decomp: small perf improvement

* update setup.py

* Update bitsandbytes/autograd/_functions.py
Co-authored-by: Aarni Koskela <akx@iki.fi>

* Update bitsandbytes/functional.py
Co-authored-by: Aarni Koskela <akx@iki.fi>

* Update bitsandbytes/functional.py
Co-authored-by: Aarni Koskela <akx@iki.fi>

* Update bitsandbytes/research/autograd/_functions.py
Co-authored-by: Aarni Koskela <akx@iki.fi>

* int8 - perf improvement for sparse decomposition inference; deprecate get_tensor_stream() in favor of new private fn

* int8 cleanup

* Ignore ruff rule ISC001 (incompatible with formatter)

* add comment

* int8 more cleanup

* Update bitsandbytes/functional.py
Co-authored-by: Aarni Koskela <akx@iki.fi>

* int8: rename / deprecate old fn signatures

* Update bitsandbytes/functional.py
Co-authored-by: Aarni Koskela <akx@iki.fi>

* type annotation

* format update

* Update bitsandbytes/research/autograd/_functions.py
Co-authored-by: Aarni Koskela <akx@iki.fi>

* cleanup

* Add comment to explain division optimization

* more cleanup

* Update bitsandbytes/functional.py
Co-authored-by: Aarni Koskela <akx@iki.fi>

* Update bitsandbytes/functional.py
Co-authored-by: Aarni Koskela <akx@iki.fi>

* Update bitsandbytes/functional.py
Co-authored-by: Aarni Koskela <akx@iki.fi>

* cleanup

* Type annotations, cleanup

* remove unused kernels; improved type annotations

* small perf optimization for single-GPU systems

* small perf optimization for single-GPU systems

* update docstrings

* Improve docs and tests

* Update docstring

* Update test

* add benchmarking script

* test cleanup: add deprecated marker, move benchmarks out

* Add int8 dequant function; misc improvements

* int8 matmul fallback for inner dims not divisible by 4

* improve register usage of kInt8VectorQuant - especially for A100/H100

* disable fail-fast for package build

* maxwell compat

* ptxas verbose

* docs update

* doc update

* backward fix

* Bugfix sparse decomp

* Int8 fix for PEFT OLoRA init

* Fix test for deprecated spmm_coo

* test improvement

* doc update

* typo

* doc cleanup

* docs

* add inference benchmark script

* Add benchmarks, doc update

---------
Co-authored-by: Aarni Koskela <akx@iki.fi>