Shanshan Shen d2f87ed9cc [Patch] Remove spec_decode.metrics patch (#1016) ...

### What this PR does / why we need it?
Remove `spec_decode.metrics` patch as this has been resolved in
https://github.com/vllm-project/vllm/pull/16983 (include in vllm
`v0.9.0`).

Returns a CUDA event recording when the copy is complete **--after
modified-->** Returns a device event (NPU Event for vllm-ascend)
recording when the copy is complete.

Signed-off-by: shen-shanshan <467638484@qq.com>

2025-06-09 15:05:11 +08:00

..

__init__.py

[Patch] Remove spec_decode.metrics patch (#1016)

2025-06-09 15:05:11 +08:00

patch_distributed.py

[BugFix]add all2all when dp_size > 1 && downgrade npu_dequant_swiglu_quant (#819)

2025-05-15 09:19:55 +08:00

patch_eagle.py

Spec decode support for V1 Engine (#874)

2025-05-23 14:25:46 +08:00

patch_minicpm.py

[Model][MiniCPM] support MiniCPM (#645)

2025-04-27 11:27:24 +08:00

patch_multi_step_worker.py

[CI] upgrade vllm to 0.8.5 (#715)

2025-04-30 09:15:50 +08:00

patch_sampler.py

Fix typo of VLLM_ASCEND_ENABLE_TOPK_OPTIMIZE (#1112)

2025-06-07 19:45:33 +08:00

patch_spec_decode_worker.py

[SpecDecode][CI] Set default values to fix spec decode and fix multicard CI (#1109)

2025-06-07 11:23:30 +08:00

patch_utils.py

[CI] Patch torch.library.infer_schema for torch 2.5 backward compatibility (#837)

2025-05-14 09:20:55 +08:00