[Feature] totaly support multi-lora support,latest xspeedgate needed (#133)

Co-authored-by: wanghao <wanghao@example.com>
2026-01-20 21:27:02 +08:00
parent 92b40628cd
commit c404af3a41
7 changed files with 701 additions and 1 deletions
--- a/vllm_kunlun/platforms/kunlun.py
+++ b/vllm_kunlun/platforms/kunlun.py
@@ -311,7 +311,10 @@ class KunlunPlatform(Platform):

    @classmethod
    def get_punica_wrapper(cls):
-        return "vllm.lora.punica_wrapper.punica_cpu.PunicaWrapperCPU"
+        ''' 
+        kunlun wrapper
+        '''
+        return "vllm_kunlun.lora.punica_wrapper.punica_kunlun.PunicaWrapperKunlun"
    
    @classmethod
    def check_if_supports_dtype(cls, torch_dtype: torch.dtype):