Python package

python-package.yml

1065 workflow runs

Test improvements (#1750)
completed successfully:
Python package#1755:Commit6a07ffepushedbymatthewdouglas
main
Sep 9, 202514m 53s
Test improvements
completed successfully:
Python package#1754:Pull request#1750openedbymatthewdouglas
test-improvements
Sep 9, 202518m 11s
Skip test on torch==2.8.0+cpu for Windows regression
completed successfully:
Python package#1753:Commit23f542cpushedbymatthewdouglas
test-improvements
Sep 9, 202517m 48s
Skip test on torch==2.8.0+cpu for Windows regression
completed successfully:
Python package#1752:Commit0d02815pushedbymatthewdouglas
test-improvements
Sep 9, 202516m 10s
Test suite improvements for MPS/XPU/HPU
completed successfully:
Python package#1751:Commit2ffbf22pushedbymatthewdouglas
test-improvements
Sep 9, 20258m 59s
Adjust 4bit test tolerance on CPU for larger blocksizes (#1749)
completed successfully:
Python package#1750:Commitd731fc4pushedbymatthewdouglas
main
Sep 8, 202518m 20s
Adjust 4bit test tolerance on CPU for larger blocksizes
completed successfully:
Python package#1749:Pull request#1749openedbymatthewdouglas
fix-cpu-4bit-tests
Sep 8, 202518m 44s
Adjust 4bit test tolerance on CPU for larger blocksizes
completed successfully:
Python package#1748:Commit918def4pushedbymatthewdouglas
fix-cpu-4bit-tests
Sep 8, 202517m 33s
4bit quantization for arbitrary nn.Parameter (#1720)
completed successfully:
Python package#1747:Commit27549fbpushedbymatthewdouglas
main
Sep 8, 202558m 22s
add support for 64 block size on 32 warp size supported amd gpus
completed successfully:
Python package#1745:Pull request#1748synchronizebyelectron271
electron271:main
Sep 6, 202510m 58s
[CUDA] Branchless NF4/FP4 kDequantizeBlockwise kernel for faster dequantization
completed successfully:
Python package#1743:Pull request#1746synchronizebyMhmd-Hisham
Mhmd-Hisham:cuda-branchless-dequantization-float32-lut
Sep 6, 202519m 49s
4bit quantization for arbitrary nn.Parameter
completed successfully:
Python package#1741:Pull request#1720synchronizebymatthewdouglas
parametrize-4bit
Sep 5, 202516m 54s
Another test gaurd for torch >= 2.5
completed successfully:
Python package#1740:Commit7fa6973pushedbymatthewdouglas
parametrize-4bit
Sep 5, 202511m 49s
Improve kDequantizeBlockwise kernel performance for NF4/FP4
completed successfully:
Python package#1739:Pull request#1747openedbymatthewdouglas
dequantize-improvement
Sep 5, 202515m 31s
Improve kDequantizeBlockwise kernel performance for NF4/FP4
completed successfully:
Python package#1738:Commit65b66b2pushedbymatthewdouglas
dequantize-improvement
Sep 5, 202517m 25s
4bit quantization for arbitrary nn.Parameter
completed successfully:
Python package#1735:Pull request#1720synchronizebymatthewdouglas
parametrize-4bit
Sep 3, 202514m 24s
Guard for torch < 2.5
completed successfully:
Python package#1734:Commit62267d7pushedbymatthewdouglas
parametrize-4bit
Sep 3, 202513m 25s
"Claude Code Review workflow"
completed successfully:
Python package#1733:Commit8823837pushedbyTitus-von-Koeller
add-claude-github-actions-1756931485147
Sep 3, 202517m 35s
"Claude PR Assistant workflow"
cancelled:
Python package#1732:Commit59b531dpushedbyTitus-von-Koeller
add-claude-github-actions-1756931485147
Sep 3, 20252s
for intel xpu case, use MatMul8bitFp even not use ipex (#1728)
cancelled:
Python package#1731:Commit39dd847pushedbyTitus-von-Koeller
add-claude-github-actions-1756931485147
Sep 3, 20254s
4bit quantization for arbitrary nn.Parameter
completed successfully:
Python package#1730:Pull request#1720synchronizebymatthewdouglas
parametrize-4bit
Sep 3, 202519m 22s
Guard for torch < 2.5
completed successfully:
Python package#1729:Commit794710dpushedbymatthewdouglas
parametrize-4bit
Sep 3, 202513m 14s
4bit quantization for arbitrary nn.Parameter
completed successfully:
Python package#1728:Pull request#1720synchronizebymatthewdouglas
parametrize-4bit
Sep 3, 202510m 59s
Fix tests
completed successfully:
Python package#1727:Commit0d5cda7pushedbymatthewdouglas
parametrize-4bit
Sep 3, 202511m 5s
4bit quantization for arbitrary nn.Parameter
cancelled:
Python package#1726:Pull request#1720synchronizebymatthewdouglas
parametrize-4bit
Sep 3, 20254m 26s

Navigation Menu

Python package

1065 workflow runs

Test improvements (#1750)

Test improvements

Skip test on torch==2.8.0+cpu for Windows regression

Skip test on torch==2.8.0+cpu for Windows regression

Test suite improvements for MPS/XPU/HPU

Adjust 4bit test tolerance on CPU for larger blocksizes (#1749)

Adjust 4bit test tolerance on CPU for larger blocksizes

Adjust 4bit test tolerance on CPU for larger blocksizes

4bit quantization for arbitrary `nn.Parameter` (#1720)

add support for 64 block size on 32 warp size supported amd gpus

[CUDA] Branchless NF4/FP4 kDequantizeBlockwise kernel for faster dequantization

4bit quantization for arbitrary `nn.Parameter`

Another test gaurd for torch >= 2.5

Improve kDequantizeBlockwise kernel performance for NF4/FP4

Improve kDequantizeBlockwise kernel performance for NF4/FP4

4bit quantization for arbitrary `nn.Parameter`

Guard for torch < 2.5

"Claude Code Review workflow"

"Claude PR Assistant workflow"

for intel xpu case, use MatMul8bitFp even not use ipex (#1728)

4bit quantization for arbitrary `nn.Parameter`

Guard for torch < 2.5

4bit quantization for arbitrary `nn.Parameter`

Fix tests

4bit quantization for arbitrary `nn.Parameter`

ProTip! You can narrow down the results and go further in time using the filter or the other filters available.

Workflow runs for Python package

1065 workflow runs

ProTip! You can narrow down the results and go further in time using the created: filter or the other filters available.

ProTip! You can narrow down the results and go further in time using the filter or the other filters available.