Skip to content

Actions: bitsandbytes-foundation/bitsandbytes

Actions

Python package

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
python-package.yml
1,077 workflow runs
1,077 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
[ROCm] Preserve float bits in SIMT DPP reduction Python package #3009: Pull request #2089 opened by liminfei-amd
Action required liminfei-amd:fix/2087-gfx1100-dpp-bitcast
liminfei-amd:fix/2087-gfx1100-dpp-bitcast
Action required
fix(optim): reject bias_correction and adam_w_mode in LAMB Python package #3008: Pull request #2088 opened by ConnorMoss02
Action required ConnorMoss02:fix/lamb-reject-unsupported-args
ConnorMoss02:fix/lamb-reject-unsupported-args
Action required
fix(optim): reject nesterov and dampening in SGD and LARS Python package #3007: Pull request #2085 opened by ConnorMoss02
Action required ConnorMoss02:fix/sgd-lars-reject-unsupported-nesterov-dampening
ConnorMoss02:fix/sgd-lars-reject-unsupported-nesterov-dampening
Action required
fix(cpu): correctly handle 1D output shape in dequantize_4bit Python package #3006: Pull request #2055 synchronize by SparshM8
1h 19m 18s SparshM8:fix/cpu-dequantize-1d-shape
SparshM8:fix/cpu-dequantize-1d-shape
1h 19m 18s
Fix Params4bit.to() leaving CPU-packed weights packed Python package #3005: Pull request #2083 opened by 22elix3r
Action required 22elix3r:fix/params4bit-to-cpu-packing
22elix3r:fix/params4bit-to-cpu-packing
Action required
Fix PytorchLARS updates when momentum is zero Python package #3004: Pull request #2082 opened by mottopanikeiku
Action required mottopanikeiku:fix/issue-2079
mottopanikeiku:fix/issue-2079
Action required
Warn once when Linear4bit computes in float32 on GPU Python package #3003: Pull request #2077 opened by thangdangjp
Action required thangdangjp:feature/warn-fp32-compute-dtype
thangdangjp:feature/warn-fp32-compute-dtype
Action required
Make AdEMAMix32bit inherit AdEMAMix so it allocates the double state buffer Python package #3002: Pull request #2076 opened by caiotheodoro
Action required caiotheodoro:ademamix32bit-double-buffer
caiotheodoro:ademamix32bit-double-buffer
Action required
chore(deps): bump softprops/action-gh-release from 3.0.2 to 3.0.3 in the actions group Python package #3001: Pull request #2074 opened by dependabot Bot
10m 35s dependabot/github_actions/actions-6512b1d693
dependabot/github_actions/actions-6512b1d693
10m 35s
chore(deps): bump softprops/action-gh-release in the actions group Python package #3000: Commit 832d3c1 pushed by dependabot Bot
9m 28s dependabot/github_actions/actions-6512b1d693
dependabot/github_actions/actions-6512b1d693
9m 28s
[ROCm] Allow explicit GPU architecture override Python package #2999: Pull request #2073 opened by LOGO127
Action required LOGO127:fix/rocm-arch-env-override
LOGO127:fix/rocm-arch-env-override
Action required
Add AVX2 path for CPU 4-bit (NF4/FP4) blockwise dequantization Python package #2998: Pull request #2072 synchronize by rachitj27
Action required rachitj27:feature/cpu-avx2-4bit-dequant
rachitj27:feature/cpu-avx2-4bit-dequant
Action required
Add AVX2 path for CPU 4-bit (NF4/FP4) blockwise dequantization Python package #2997: Pull request #2072 opened by rachitj27
Action required rachitj27:feature/cpu-avx2-4bit-dequant
rachitj27:feature/cpu-avx2-4bit-dequant
Action required
Fix CPU Hub kernel backend selection Python package #2996: Pull request #2065 synchronize by SiluPanda
12m 24s SiluPanda:fix/issue-2064
SiluPanda:fix/issue-2064
12m 24s
fix(cuda): chunk 4-bit quant/dequant launches above INT32_MAX elements (#1785) Python package #2995: Pull request #2071 opened by Anai-Guo
Action required Anai-Guo:fix/4bit-chunk-int32-launch
Anai-Guo:fix/4bit-chunk-int32-launch
Action required
Feat/mps gemm 4bit bf16 Python package #2994: Pull request #2070 opened by eaglstun
12m 41s eaglstun:feat/mps-gemm-4bit-bf16
eaglstun:feat/mps-gemm-4bit-bf16
12m 41s
fix(optim): fall back to non-paged state on all devices without managed memory Python package #2993: Pull request #2069 synchronize by egeozkoc
Action required egeozkoc:fix/paged-optimizer-non-cuda
egeozkoc:fix/paged-optimizer-non-cuda
Action required
fix(optim): fall back to non-paged state on all devices without managed memory Python package #2992: Pull request #2069 opened by egeozkoc
9m 16s egeozkoc:fix/paged-optimizer-non-cuda
egeozkoc:fix/paged-optimizer-non-cuda
9m 16s
(docs) Fix RTX Spark PyTorch index URL (#2068) Python package #2991: Commit 8336490 pushed by matthewdouglas
10m 31s main
main
10m 31s
(docs) Fix RTX Spark PyTorch index URL Python package #2990: Commit b8e8a38 pushed by matthewdouglas
14m 32s doc-update-rtxspark-index-url
doc-update-rtxspark-index-url
14m 32s
Fix CPU Hub kernel backend selection Python package #2988: Pull request #2065 opened by SiluPanda
20m 26s SiluPanda:fix/issue-2064
SiluPanda:fix/issue-2064
20m 26s
fix(optim): mark unused amsgrad/optim_bits as deprecated in 8-bit Adam classes Python package #2987: Pull request #2063 opened by kritikagarg
Action required kritikagarg:fix/adam8bit-deprecated-params-docs
kritikagarg:fix/adam8bit-deprecated-params-docs
Action required
Fix N-D input handling in default backend int8 ops Python package #2986: Pull request #2062 opened by 2sumtech
Action required 2sumtech:fix/default-int8-ndim
2sumtech:fix/default-int8-ndim
Action required
Fix CPU dequantize_4bit output shape for 1-D inputs Python package #2985: Pull request #2048 synchronize by 2sumtech
12m 44s 2sumtech:fix/dequantize-4bit-1d-shape
2sumtech:fix/dequantize-4bit-1d-shape
12m 44s
Bump dev version Python package #2984: Commit 2a7fd00 pushed by matthewdouglas
3m 32s main
main
3m 32s