Update handler.py

handler.py

@@ -36,6 +36,12 @@ class EndpointHandler:
         ).to("cuda")
 
         with torch.backends.cuda.sdp_kernel(enable_flash=True, enable_math=False, enable_mem_efficient=False):
-            outputs = self.model.generate(**inputs)
+            outputs = self.model.generate(
+                input_ids=inputs,
+                max_new_tokens=64,
+                use_cache=True,
+                temperature=1.5,
+                min_p=0.1,
+            )
 
         return self.tokenizer.decode(outputs[0], skip_special_tokens=True)