feat: add initial AutoGPTQ backend implementation

2025-06-19 07:28:08 +00:00 · 2023-08-07 22:39:10 +02:00
parent 91d49cfe9f
commit a843e64fc2
37 changed files with 660 additions and 148 deletions
--- a/extra/grpc/huggingface/huggingface.py
+++ b/extra/grpc/huggingface/huggingface.py
@ -18,7 +18,6 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
        return backend_pb2.Reply(message=bytes("OK", 'utf-8'))
    def LoadModel(self, request, context):
        model_name = request.Model
-        model_name = os.path.basename(model_name)
        try:
            self.model = SentenceTransformer(model_name)
        except Exception as err: