Matthias GehreandAndreas Karatzas 382bbd5144 [ROCm][Kernel] Add HybridW4A16LinearKernel: Triton prefill + HIP skinny decode (#40977) ...

Signed-off-by: Matthias Gehre <matthias.gehre@amd.com>
Co-authored-by: Andreas Karatzas <akaratza@amd.com>

2026-07-13 20:22:00 -07:00

..

kernels

[ROCm][Kernel] Add HybridW4A16LinearKernel: Triton prefill + HIP skinny decode (#40977)

2026-07-13 20:22:00 -07:00

layers

[ROCm][MiniMax-M2] Dispatch fused QK-norm + AllReduce via AITER (#44849)

2026-07-14 03:10:16 +00:00

model_loader

fix: correct load_weights track logic and enable weight integrity for… (#41811)

2026-07-10 14:08:20 +00:00

models

[Feat] Add Support for BertForMaskedLM to vLLM (#48463)

2026-07-13 20:56:25 +00:00

offloader

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

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

warmup

Add VLLM_FLASHINFER_AUTOTUNE_SKIP_OPS and skip CuTeDSL fp4_gemm autotuning by default (#48268)

2026-07-10 18:13:08 -07:00

__init__.py

[Platform] Deprecate seed_everything (#31659)

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

custom_op.py

Add ability to replace oot ops when using lora (#37181)

2026-03-16 18:04:15 -07:00

parameter.py

Speed up docs build (#44635)

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

utils.py

Remove more unnecessary load_weights methods (#47058)

2026-06-30 15:22:16 +01:00