Skip to content

Actions: bitsandbytes-foundation/bitsandbytes

Actions

Python package

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
python-package.yml
1,055 workflow runs
1,055 workflow runs

Filter by Event

⌥ + click/return to exclude

Filter by Status

⌥ + click/return to exclude

Filter by Branch

⌥ + click/return to exclude

Filter by Actor

⌥ + click/return to exclude
build: cleanup unused metal build code Python package #2855: Pull request #2004 opened by matthewdouglas
11m 37s metal-cleanup
metal-cleanup
11m 37s
build: cleanup unused metal build code Python package #2854: Commit 5e199ef pushed by matthewdouglas
12m 54s metal-cleanup
metal-cleanup
12m 54s
Deprecate igemm, batched_igemm, check_matmul APIs (#2003) Python package #2853: Commit 11822cc pushed by matthewdouglas
8m 46s main
main
8m 46s
Deprecate igemm, batched_igemm, check_matmul APIs Python package #2852: Pull request #2003 opened by matthewdouglas
7m 22s igemm-deprecations
igemm-deprecations
7m 22s
Deprecate igemm, batched_igemm, check_matmul APIs Python package #2851: Commit 5610b1b pushed by matthewdouglas
8m 44s igemm-deprecations
igemm-deprecations
8m 44s
XPU: Add oneAPI 2026 build; package both 2025 and 2026 variants Python package #2850: Pull request #2002 opened by matthewdouglas
8m 6s xpu2026
xpu2026
8m 6s
XPU: Add oneAPI 2026 build; package both 2025 and 2026 variants Python package #2849: Commit 0e77874 pushed by matthewdouglas
9m 20s xpu2026
xpu2026
9m 20s
Fix Lion to use decoupled weight decay in the Triton 32-bit kernel Python package #2848: Pull request #2001 synchronize by eaglstun
8m 16s eaglstun:fix/lion-triton-decoupled-wd-v2
eaglstun:fix/lion-triton-decoupled-wd-v2
8m 16s
fix(parametrize): cache-gate hooks leak the global parametrize cache under activation checkpointing (all dequantized params retained) Python package #2846: Pull request #1999 synchronize by pjordanandrsn
9m 12s pjordanandrsn:fix/parametrize-cache-leak
pjordanandrsn:fix/parametrize-cache-leak
9m 12s
fix(optim): honor user-provided max_unorm in LAMB (+ 8-bit arg guards) Python package #2844: Pull request #1998 opened by egeozkoc
7m 20s egeozkoc:fix/issue-1261
egeozkoc:fix/issue-1261
7m 20s
Add BF16 support for int8_vectorwise_quant / LLM.int8 activation quant Python package #2842: Pull request #1985 synchronize by kru2710shna
6m 49s kru2710shna:feature/int8-bf16-vectorwise-quant
kru2710shna:feature/int8-bf16-vectorwise-quant
6m 49s
ci: bump Linux ROCm support to 7.2.4 (#1997) Python package #2841: Commit 7ec290a pushed by matthewdouglas
8m 33s main
main
8m 33s
ci: bump Linux ROCm support to 7.2.4 Python package #2840: Pull request #1997 opened by sstamenk
8m 30s sstamenk:feat/rocm-7.2.4
sstamenk:feat/rocm-7.2.4
8m 30s
Handle non-contiguous inputs in CPU blockwise quant/dequant (#1996) Python package #2839: Commit 771db27 pushed by matthewdouglas
7m 48s main
main
7m 48s
Handle non-contiguous inputs in CPU blockwise quantize/dequantize Python package #2837: Pull request #1996 synchronize by egeozkoc
7m 48s egeozkoc:feature/cpu-noncontiguous-blockwise
egeozkoc:feature/cpu-noncontiguous-blockwise
7m 48s
Enable non-contiguous quantize_4bit tests for MPS backend (#1994) Python package #2835: Commit e56da9e pushed by matthewdouglas
8m 15s main
main
8m 15s
Enable non-contiguous quantize_4bit tests for the MPS backend Python package #2832: Pull request #1994 opened by egeozkoc
8m 51s egeozkoc:feature/mps-noncontiguous-quant
egeozkoc:feature/mps-noncontiguous-quant
8m 51s
Fix Lion to use decoupled weight decay (default + CUDA 32-bit) (#1993) Python package #2831: Commit ab9709a pushed by matthewdouglas
8m 2s main
main
8m 2s
Fix Lion to use decoupled weight decay (default + CUDA 32-bit) Python package #2830: Pull request #1993 synchronize by matthewdouglas
9m 26s eaglstun:fix/issue-1992
eaglstun:fix/issue-1992
9m 26s
Fix Lion to use decoupled weight decay (default + CUDA 32-bit) Python package #2829: Pull request #1993 opened by eaglstun
7m 25s eaglstun:fix/issue-1992
eaglstun:fix/issue-1992
7m 25s
Add BF16 support for int8_vectorwise_quant / LLM.int8 activation quant Python package #2828: Pull request #1985 synchronize by kru2710shna
7m 5s kru2710shna:feature/int8-bf16-vectorwise-quant
kru2710shna:feature/int8-bf16-vectorwise-quant
7m 5s
Fix compat with sckit-build-core 1.0.0 (#1989) Python package #2827: Commit 777c145 pushed by matthewdouglas
8m 11s main
main
8m 11s
Fix compat with scikit-build-core 1.0.0 Python package #2826: Pull request #1989 opened by matthewdouglas
7m 54s sckit-build-core-v1-compat
sckit-build-core-v1-compat
7m 54s
Fix compat with sckit-build-core 1.0.0 Python package #2825: Commit aad7f8b pushed by matthewdouglas
7m 15s sckit-build-core-v1-compat
sckit-build-core-v1-compat
7m 15s
[ROCm] Fused 4-bit SIMT GEMM (#1979) Python package #2824: Commit 6e50f39 pushed by matthewdouglas
7m 36s main
main
7m 36s