Skip to content

Actions: bitsandbytes-foundation/bitsandbytes

Actions

PR Tests

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
tests-pr.yml
171 workflow runs
171 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 target specific option mismatch during OpenMP loop inlining under GCC 16 PR Tests #280: Pull request #1954 opened by Soddentrough
Action required Soddentrough:fix-gcc16-openmp-inlining
Soddentrough:fix-gcc16-openmp-inlining
Action required
Reduce Python CPU overhead and improve validation messages PR Tests #279: Pull request #1953 opened by matthewdouglas
9m 53s reduce-cpu-overhead
reduce-cpu-overhead
9m 53s
4bit GEMM fix: per-device cudaFuncSetAttribute cache PR Tests #278: Pull request #1952 opened by matthewdouglas
25m 56s gemm4bit-fix-setattribute
gemm4bit-fix-setattribute
25m 56s
[CUDA] New 4bit GEMM kernels for inference PR Tests #276: Pull request #1949 synchronize by matthewdouglas
26m 24s gemm4bit
gemm4bit
26m 24s
[CUDA] New 4bit GEMM kernels for inference PR Tests #275: Pull request #1949 synchronize by matthewdouglas
27m 37s gemm4bit
gemm4bit
27m 37s
[CUDA] New 4bit GEMM kernels for inference PR Tests #274: Pull request #1949 synchronize by matthewdouglas
1h 26m 2s gemm4bit
gemm4bit
1h 26m 2s
CI: Update to test with torch 2.12.0 and nightly PR Tests #273: Pull request #1948 opened by matthewdouglas
27m 20s ci-torch212-update
ci-torch212-update
27m 20s
chore: bump minimum PyTorch version from 2.3 to 2.4 PR Tests #272: Pull request #1926 synchronize by matthewdouglas
45m 40s chore/bump-min-torch-2.4
chore/bump-min-torch-2.4
45m 40s
chore: bump minimum PyTorch version from 2.3 to 2.4 PR Tests #271: Pull request #1926 synchronize by matthewdouglas
42m 36s chore/bump-min-torch-2.4
chore/bump-min-torch-2.4
42m 36s
CI Updates PR Tests #270: Pull request #1947 synchronize by matthewdouglas
34m 27s fix-torch-nightly-ci
fix-torch-nightly-ci
34m 27s
CI Updates PR Tests #269: Pull request #1947 opened by matthewdouglas
24m 38s fix-torch-nightly-ci
fix-torch-nightly-ci
24m 38s
refactor: Rewrite assert statements as exceptions PR Tests #268: Pull request #1931 synchronize by rapsealk
Action required rapsealk:rewrite-asserts-as-exceptions
rapsealk:rewrite-asserts-as-exceptions
Action required
Add PBF4: log-polar 4-bit weight format alongside fp4/nf4 PR Tests #267: Pull request #1943 opened by kgrama
Action required kgrama:logtables
kgrama:logtables
Action required
Fix gemv_4bit bfloat16 correctness on Intel Arc A-series (Alchemist) GPUs PR Tests #266: Pull request #1942 opened by jiqing-feng
28m 56s jiqing-feng:kernel
jiqing-feng:kernel
28m 56s
refactor: Rewrite assert statements as exceptions PR Tests #265: Pull request #1931 synchronize by rapsealk
Action required rapsealk:rewrite-asserts-as-exceptions
rapsealk:rewrite-asserts-as-exceptions
Action required
fix: replace all instances of deprecated torch._check_is_size with torch._check PR Tests #264: Pull request #1940 opened by neil-the-nowledgeable
27m 51s neil-the-nowledgeable:fix/check-is-size-deprecation
neil-the-nowledgeable:fix/check-is-size-deprecation
27m 51s
Add sm87 to aarch64 CUDA wheel targets (Jetson Orin) PR Tests #262: Pull request #1939 opened by neil-the-nowledgeable
Action required neil-the-nowledgeable:fix/issue-1930
neil-the-nowledgeable:fix/issue-1930
Action required
fix: prevent F.linear from saving dequantized weights in MatMul4Bit/MatMul8bitLt to save ~13GB VRAM and prevent OOM errors PR Tests #260: Pull request #1935 synchronize by butterwecksolutions
Action required butterwecksolutions:main
butterwecksolutions:main
Action required
fix: prevent F.linear from saving dequantized weights in MatMul4Bit/MatMul8bitLt to save ~13GB VRAM and prevent OOM errors PR Tests #259: Pull request #1935 synchronize by butterwecksolutions
Action required butterwecksolutions:main
butterwecksolutions:main
Action required
fix: prevent F.linear from saving dequantized weights in MatMul4Bit/MatMul8bitLt to save ~13GB VRAM and prevent OOM errors PR Tests #258: Pull request #1935 opened by butterwecksolutions
Action required butterwecksolutions:main
butterwecksolutions:main
Action required
[ROCm] Try multiple hipBLASLt heuristic algos and fall back to fp32 for unsupported int8 shapes PR Tests #257: Pull request #1934 opened by Abdennacer-Badaoui
27m 28s Abdennacer-Badaoui:fix-hipblast-algos
Abdennacer-Badaoui:fix-hipblast-algos
27m 28s
[ROCm] Optimize kgemm_4bit_inference_naive for ROCm, use it for batch sizes other than 1 PR Tests #255: Pull request #1920 synchronize by sstamenk
28m 11s sstamenk:rocm-4bit-kernel-optimization
sstamenk:rocm-4bit-kernel-optimization
28m 11s
feat: add anonymous feature-usage telemetry PR Tests #252: Pull request #1928 opened by Titus-von-Koeller
26m 41s feat/telemetry
feat/telemetry
26m 41s
chore: bump minimum PyTorch version from 2.3 to 2.4 PR Tests #251: Pull request #1926 synchronize by Titus-von-Koeller
23m 27s chore/bump-min-torch-2.4
chore/bump-min-torch-2.4
23m 27s
Enable XPU for blockwise quantization and FSDP tests PR Tests #250: Pull request #1921 synchronize by jiqing-feng
3m 35s jiqing-feng:clean_tests
jiqing-feng:clean_tests
3m 35s