vllm-npu-plugin/vllm_npu at f54533fba760ff080dc4c22dd374728a41f506d2 - vllm-npu-plugin - Gitea: Git with a cup of tea

zzh/vllm-npu-plugin

mirror of https://github.com/handsomezhuzhu/vllm-npu-plugin.git synced 2026-02-20 11:42:30 +00:00

Files

History

handsomezhuzhu f54533fba7 fix: use 4D mask (1, 1, S, S) for BSND layout in npu_fusion_attention

2026-02-10 20:57:52 +08:00

..

fix: use 4D mask (1, 1, S, S) for BSND layout in npu_fusion_attention

2026-02-10 20:57:52 +08:00

feat: initial vllm-npu-plugin for Ascend NPU adaptation

2026-02-10 11:06:01 +08:00

feat: initial vllm-npu-plugin for Ascend NPU adaptation

2026-02-10 11:06:01 +08:00

fix: add initialize_cache method to NPU worker

2026-02-10 19:42:32 +08:00

__init__.py

feat: add CUDA-to-NPU monkey patches for GPUModelRunner compatibility

2026-02-10 19:09:14 +08:00

cuda_compat.py

feat: add CUDA-to-NPU monkey patches for GPUModelRunner compatibility

2026-02-10 19:09:14 +08:00

platform.py

feat: initial vllm-npu-plugin for Ascend NPU adaptation

2026-02-10 11:06:01 +08:00