Wentao Ye fba31f31a9 [GLM5.3 Perf] Switch to fp8 kv cache by default for GLM, 2.3%~5.5% E2E Throughput Improvement (#60140) ...

Signed-off-by: yewentao256 <zhyanwentao@126.com>

2026-10-06 12:54:20 -04:00

..

determinism

[Bugfix][Determinism] Preserve output dtype in batch-invariant mean (#59106)

2026-10-04 17:07:46 +00:00

hw_agnostic

[CI] Ignore ruff D209, rejoin the docstrings it split, and silence incompatible-rule warnings (#57212)

2026-09-16 10:57:24 -07:00

kernels

[ROCm][Perf] W4A16: pad gfx11 weight and activation strides (#56301)

2026-10-06 08:22:33 -05:00

layers

[Quantization] Layer re-quantization for linear layers through online quantization API (MXFP8 -> FP8 PTPC showcase) (#55684)

2026-10-06 11:38:18 -05:00

model_loader

[Quantization] Layer re-quantization for linear layers through online quantization API (MXFP8 -> FP8 PTPC showcase) (#55684)

2026-10-06 11:38:18 -05:00

models

[GLM5.3 Perf] Switch to fp8 kv cache by default for GLM, 2.3%~5.5% E2E Throughput Improvement (#60140)

2026-10-06 12:54:20 -04:00

offloader

[MRV2] Release weight offloader on shutdown (#57834)

2026-09-22 15:01:20 -04:00

warmup

[Build] Migrate vendored DeepGEMM from pybind to TORCH_LIBRARY (abi3) (#48962)

2026-10-06 08:39:04 -07:00

__init__.py

[Platform] Deprecate seed_everything (#31659)

2026-01-04 18:34:04 -08:00

custom_op.py

Add pydocstyle to the ruff rules (#52136)

2026-09-16 16:44:29 +00:00

parameter.py

Add pydocstyle to the ruff rules (#52136)

2026-09-16 16:44:29 +00:00

utils.py

[CI] Ignore ruff D209, rejoin the docstrings it split, and silence incompatible-rule warnings (#57212)

2026-09-16 10:57:24 -07:00