vLLM/vllm/models at patches - vLLM - iHore Software Repository

iHore/vLLM

Files

T

History

MOVZX 49d708e158 Merge branch 'main' into patches

Signed-off-by: MOVZX <movzx@yahoo.com>

2026-10-18 18:09:11 +00:00

..

[Misc] Consolidate csrc shared headers and prune dead code (#60721 )

2026-10-12 09:46:22 -07:00

[ROCm][DSv4] Overlap the unfused shared expert with AITER MegaMoE (#61819 )

2026-10-18 22:39:38 +08:00

[Bugfix][ROCm][GLM-5.2] Stop forcing skip_final_all_reduce on AMD DSA model path (#58589 )

2026-10-16 21:38:55 +08:00

[Model] DeepSeek-V4/V4.1 VL: stream language-model weights instead of sorting the checkpoint (#58857 )

2026-10-17 09:37:56 +00:00

[Chore] Use Transformers v5 names and drop redundant processor use_fast (#58550 )

2026-09-24 14:14:10 +00:00

[Perf][ROCm][GLM-5.3-Flash] AITER prefill indexer topk (#60753 )

2026-10-14 13:48:56 -05:00

[Bugfix][RL] Load attention sinks through weight_loader for layerwise reload (#57797 )

2026-10-14 10:17:58 +00:00

[Chore] Use upstream Transformers InklingProcessor (#61883 )

2026-10-17 01:04:54 +00:00

Merge branch 'main' into patches

2026-10-18 16:04:36 +07:00

[Bugfix][ROCm][MiniMax-M3] Reject unbuilt index-head counts in the AITER indexer (#60990 )

2026-10-14 14:13:48 -05:00

Merge branch 'main' into patches

2026-10-18 18:09:11 +00:00

__init__.py

[Model Refactoring] Migrate DeepSeek V4 to vllm/models/ [1/N] (#43004 )

2026-05-18 19:50:02 -07:00