Fangzhou Ai 1044d19998 [ROCm][DSV4] Preserve native MXFP4 TP8 shard allocation (#51473) ...

Signed-off-by: Fangzhou Ai <fangzhou.ai@amd.com>

2026-08-10 23:57:27 -07:00

..

kernels

[CPU][Zen] Route BF16 MoE inference through zentorch on AMD (#44201)

2026-08-10 23:20:05 -07:00

layers

[ROCm][DSV4] Preserve native MXFP4 TP8 shard allocation (#51473)

2026-08-10 23:57:27 -07:00

model_loader

Preserve revision pins in secondary artifact loaders (#51446)

2026-08-11 04:33:10 +00:00

models

Fix docs on main (#51773)

2026-08-10 23:52:04 -07:00

offloader

Revert "[Platform] Replace torch.cuda.Event with torch.Event (#47140)" (#47668)

2026-07-06 14:18:33 +01:00

warmup

Bump Flashinfer version to 0.6.16.post3 (#50892)

2026-08-09 00:54:07 -07:00

__init__.py

[Platform] Deprecate seed_everything (#31659)

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

custom_op.py

[Bugfix] Reject contradictory custom-op directives (#49134)

2026-07-26 08:42:14 -04:00

parameter.py

Speed up docs build (#44635)

2026-06-05 14:51:44 +00:00

utils.py

[Weight processing] Copy over new_data attributes in replace_parameter (#49601)

2026-08-06 13:03:37 -04:00