vllmellm 1a19e9cd87 [Bugfix][ROCm]Fix Qwen3-Next-80B-A3B-Thinking inference and optimize non-standard block size (544) support under rocm_atten (#31380) ...

Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>

2026-01-09 19:28:02 +08:00

..

backends

[Chore] Migrate V0 attention utils (#31891)

2026-01-07 13:44:36 +00:00

layers

[MM Encoder]: Make MMEncoderAttention's scale takes effect properly (#31950)

2026-01-08 02:33:48 -08:00

ops

[Bugfix][ROCm]Fix Qwen3-Next-80B-A3B-Thinking inference and optimize non-standard block size (544) support under rocm_atten (#31380)

2026-01-09 19:28:02 +08:00

utils

[Misc] Fix Current vLLM config is not set. warnings, assert to avoid issues in the future (#31747)

2026-01-08 15:20:49 -08:00

__init__.py

[Attention] Remove imports from vllm/attention/__init__.py (#29342)

2025-11-26 10:53:15 -07:00

layer.py

[MoE] Fix output_shape calculation in Attention layer to handle 3D query inputs (#31596)

2026-01-02 15:46:23 +00:00

selector.py

[Platform] Refactor Platform attention backend selection to avoid breakpoint for OOT platform (#30212)

2025-12-15 17:36:07 +00:00