Skip to content

Actions: bitsandbytes-foundation/bitsandbytes

Actions

Python package

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
python-package.yml
1,204 workflow runs
1,204 workflow runs

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

Remove legacy sparse functionality Python package #2363: Commit 229217a pushed by matthewdouglas
8m 26s remove-sparse
remove-sparse
8m 26s
Remove unused kernels (#1880) Python package #2362: Commit d5cb0f2 pushed by matthewdouglas
9m 52s main
main
9m 52s
Remove non-blockwise 8-bit optimizer and legacy quantization CUDA/HIP kernels Python package #2361: Pull request #1880 opened by matthewdouglas
5m 25s remove-unused-kernels
remove-unused-kernels
5m 25s
Remove unused kernels Python package #2360: Commit 6018d9a pushed by matthewdouglas
8m 50s remove-unused-kernels
remove-unused-kernels
8m 50s
Remove deprecated APIs: research module, non-blockwise optimizers, an… Python package #2359: Commit c27dc58 pushed by matthewdouglas
9m 37s main
main
9m 37s
bench: Add comprehensive VLM benchmark + fix CUDA graph timing method… Python package #2358: Commit 14ee2e9 pushed by TimDettmers
7m 44s feature/kbit-gemv-v8
feature/kbit-gemv-v8
7m 44s
refactor: Remove random sign flips from Hadamard rotation Python package #2357: Commit fcfca9f pushed by TimDettmers
8m 6s feature/kbit-gemv-v8
feature/kbit-gemv-v8
8m 6s
style: Apply pre-commit formatting (ruff, clang-format) Python package #2356: Commit 28fa6c2 pushed by TimDettmers
6m 48s feature/kbit-gemv-v8
feature/kbit-gemv-v8
6m 48s
Unify CUDA and HIP kernel sources via compat.cuh portability layer Python package #2355: Pull request #1877 synchronize by Abdennacer-Badaoui
9m 31s Abdennacer-Badaoui:merge-cuda-hip
Abdennacer-Badaoui:merge-cuda-hip
9m 31s
perf: Remove unnecessary workspace allocation from NVFP4 GEMM Python package #2354: Commit 0ba4950 pushed by TimDettmers
7m 8s feature/qutlass-nvfp4-gemm
feature/qutlass-nvfp4-gemm
7m 8s
perf: Make fused quantize CUDA-graph-capturable Python package #2353: Commit 9d2f452 pushed by TimDettmers
7m 5s feature/qutlass-nvfp4-gemm
feature/qutlass-nvfp4-gemm
7m 5s
feat: CPU-to-GPU streaming quantization for minimal peak GPU memory Python package #2352: Commit 9a13bae pushed by TimDettmers
6m 55s QLORA-2
QLORA-2
6m 55s
feat: Per-rank model loading for pipeline parallelism Python package #2351: Commit 814e665 pushed by TimDettmers
9m 0s QLORA-2
QLORA-2
9m 0s
style: Apply ruff-format and clang-format from pre-commit Python package #2350: Commit 6a2043e pushed by TimDettmers
7m 50s feature/qutlass-nvfp4-gemm
feature/qutlass-nvfp4-gemm
7m 50s
docs: Update benchmarks, architecture docs, and license credits for C… Python package #2349: Commit c5470b5 pushed by TimDettmers
1m 10s feature/qutlass-nvfp4-gemm
feature/qutlass-nvfp4-gemm
1m 10s
test: Add scale reorder round-trip and large-batch GEMM tests Python package #2348: Commit 2b5d8fb pushed by TimDettmers
2m 26s feature/qutlass-nvfp4-gemm
feature/qutlass-nvfp4-gemm
2m 26s
perf: Move scale reordering from GEMM dispatch to quantize time Python package #2347: Commit 19fd735 pushed by TimDettmers
1m 58s feature/qutlass-nvfp4-gemm
feature/qutlass-nvfp4-gemm
1m 58s
feat: Wire CUTLASS NVFP4 GEMM into Python dispatch layer Python package #2346: Commit b54aa5b pushed by TimDettmers
5m 57s feature/qutlass-nvfp4-gemm
feature/qutlass-nvfp4-gemm
5m 57s
feat: Add pipeline training example and fix first-stage integer input… Python package #2345: Commit 20d2b7e pushed by TimDettmers
6m 51s QLORA-2
QLORA-2
6m 51s
fix: Add missing cutlass/util/device_memory.h include Python package #2344: Commit 600082b pushed by TimDettmers
3m 18s feature/qutlass-nvfp4-gemm
feature/qutlass-nvfp4-gemm
3m 18s
build: CMake integration for CUTLASS NVFP4 GEMM Python package #2343: Commit b7d7e84 pushed by TimDettmers
1m 11s feature/qutlass-nvfp4-gemm
feature/qutlass-nvfp4-gemm
1m 11s
feat: Add CPU offload, Alpaca dataset, and benchmarking to training s… Python package #2342: Commit 7b5b36d pushed by TimDettmers
6m 41s QLORA-2
QLORA-2
6m 41s
docs: Update benchmarks and implementation guide with final performance Python package #2341: Commit 4f1c7cd pushed by TimDettmers
8m 31s NVFP4
NVFP4
8m 31s
style: Apply clang-format to NVFP4 GEMM kernel Python package #2340: Commit d66a5be pushed by TimDettmers
2m 39s NVFP4
NVFP4
2m 39s
perf: Two-tier split-K heuristic for better occupancy tuning Python package #2339: Commit ae46271 pushed by TimDettmers
1m 7s NVFP4
NVFP4
1m 7s