pavelzak 7ea4b40954 [Hardware][NVIDIA] Add GB10 fused-MoE fp8 tuning configs (E=256, E=512) (#52502)

2026-08-16 20:04:15 -07:00

..

hw_agnostic

Hardware-agnostic model definition via HF transformer backend (1/N) (#49458)

2026-08-13 11:21:50 +01:00

kernels

[Refactor] Remove dead code for quantization (#52221)

2026-08-16 13:10:16 -04:00

layers

[Hardware][NVIDIA] Add GB10 fused-MoE fp8 tuning configs (E=256, E=512) (#52502)

2026-08-16 20:04:15 -07:00

model_loader

[Bugfix] Avoid repeated layerwise reload warning scans (#49505)

2026-08-12 13:53:50 +08:00

models

[Bugfix][Multimodal] Keep Gemma 4 video frame counts on CPU (#52441)

2026-08-16 09:17:04 +00:00

offloader

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

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

warmup

[Kernel] Add B12X dense linear backends (#52016)

2026-08-13 19:48:33 -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