Skip to content

Actions: bitsandbytes-foundation/bitsandbytes

Actions

Python package

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
python-package.yml
1,061 workflow runs
1,061 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
fix(cpu): correctly handle 1D output shape in dequantize_4bit Python package #2969: Pull request #2055 opened by SparshM8
Action required SparshM8:fix/cpu-dequantize-1d-shape
SparshM8:fix/cpu-dequantize-1d-shape
Action required
Fuse nested 4-bit scale reconstruction on SM103 Python package #2968: Pull request #2051 synchronize by heiheiha798
Action required heiheiha798:b300/sm103-nested-4bit-dequant
heiheiha798:b300/sm103-nested-4bit-dequant
Action required
Fuse nested 4-bit scale reconstruction on SM103 Python package #2967: Pull request #2051 synchronize by heiheiha798
Action required heiheiha798:b300/sm103-nested-4bit-dequant
heiheiha798:b300/sm103-nested-4bit-dequant
Action required
Tune SM103 fused 4-bit GEMM SIMT dispatch Python package #2966: Pull request #2054 opened by heiheiha798
Action required heiheiha798:b300/sm103-4bit-simt-mma-dispatch
heiheiha798:b300/sm103-4bit-simt-mma-dispatch
Action required
Remove redundant CUB barriers from blockwise dequantization Python package #2965: Pull request #2053 opened by heiheiha798
Action required heiheiha798:b300/dequantize-blockwise-cub-barriers
heiheiha798:b300/dequantize-blockwise-cub-barriers
Action required
Respect the current CUDA stream in blockwise quantization Python package #2964: Pull request #2052 opened by heiheiha798
Action required heiheiha798:b300/quantize-blockwise-current-stream
heiheiha798:b300/quantize-blockwise-current-stream
Action required
Fuse nested 4-bit scale reconstruction on SM103 Python package #2963: Pull request #2051 opened by heiheiha798
Action required heiheiha798:b300/sm103-nested-4bit-dequant
heiheiha798:b300/sm103-nested-4bit-dequant
Action required
Use the current CUDA stream for non-paged optimizer updates Python package #2962: Pull request #2050 opened by heiheiha798
Action required heiheiha798:b300/nonpaged-optimizer-current-stream
heiheiha798:b300/nonpaged-optimizer-current-stream
Action required
ci: expand ROCm architecture coverage Python package #2961: Pull request #2046 synchronize by sstamenk
Action required sstamenk:feature/rocm-7.14-rdna1
sstamenk:feature/rocm-7.14-rdna1
Action required
Fix CPU dequantize_4bit output shape for 1-D inputs Python package #2959: Pull request #2048 opened by 2sumtech
Action required 2sumtech:fix/dequantize-4bit-1d-shape
2sumtech:fix/dequantize-4bit-1d-shape
Action required
ci: expand ROCm architecture coverage Python package #2958: Pull request #2046 synchronize by sstamenk
9m 34s sstamenk:feature/rocm-7.14-rdna1
sstamenk:feature/rocm-7.14-rdna1
9m 34s
ci: expand ROCm architecture coverage Python package #2957: Pull request #2046 synchronize by sstamenk
Action required sstamenk:feature/rocm-7.14-rdna1
sstamenk:feature/rocm-7.14-rdna1
Action required
ci: expand ROCm architecture coverage Python package #2956: Pull request #2046 opened by sstamenk
Action required sstamenk:feature/rocm-7.14-rdna1
sstamenk:feature/rocm-7.14-rdna1
Action required
[ROCm] Rework ROCm build to use ROCm version instead of HIP version Python package #2955: Pull request #1888 synchronize by sstamenk
Action required sstamenk:fix/rocm-build-rework
sstamenk:fix/rocm-build-rework
Action required
chore(deps): bump the actions group across 1 directory with 2 updates… Python package #2954: Commit 95f9af3 pushed by matthewdouglas
12m 41s main
main
12m 41s
Fix mixed INT8 FakeTensor output metadata Python package #2953: Pull request #2044 opened by tandede
Action required tandede:fix/int8-mixed-fake-metadata
tandede:fix/int8-mixed-fake-metadata
Action required
Add sm121 (GB10) to the wide_n_simt 4-bit GEMM dispatch rule (#2039) Python package #2952: Commit d2d08f6 pushed by matthewdouglas
9m 18s main
main
9m 18s
Mask the absmax store in the Triton 4-bit quantize kernels (#2043) Python package #2951: Commit d8ccdd9 pushed by matthewdouglas
5m 26s main
main
5m 26s
Mask the absmax store in the Triton 4-bit quantize kernels Python package #2950: Pull request #2043 synchronize by truong-v
8m 11s truong-v:fix/triton-4bit-absmax-store-mask
truong-v:fix/triton-4bit-absmax-store-mask
8m 11s
chore(deps): bump the actions group across 1 directory with 2 updates Python package #2949: Pull request #2036 synchronize by dependabot Bot
47s dependabot/github_actions/actions-a4d04eb940
dependabot/github_actions/actions-a4d04eb940
47s
chore(deps): bump the actions group across 1 directory with 2 updates Python package #2948: Commit f7ec8e2 pushed by dependabot Bot
7m 19s dependabot/github_actions/actions-a4d04eb940
dependabot/github_actions/actions-a4d04eb940
7m 19s
Mask the absmax store in the Triton 4-bit quantize kernels Python package #2947: Pull request #2043 opened by truong-v
22m 27s truong-v:fix/triton-4bit-absmax-store-mask
truong-v:fix/triton-4bit-absmax-store-mask
22m 27s
[ROCm] Rework ROCm build to use ROCm version instead of HIP version Python package #2946: Pull request #1888 synchronize by sstamenk
36s sstamenk:fix/rocm-build-rework
sstamenk:fix/rocm-build-rework
36s
[ROCm] Rework ROCm build to use ROCm version instead of HIP version Python package #2945: Pull request #1888 synchronize by sstamenk
Action required sstamenk:fix/rocm-build-rework
sstamenk:fix/rocm-build-rework
Action required
[ROCm] Rework ROCm build to use ROCm version instead of HIP version Python package #2944: Pull request #1888 synchronize by sstamenk
Action required sstamenk:fix/rocm-build-rework
sstamenk:fix/rocm-build-rework
Action required