auto-assign gpu memory with --auto-devices alone

server.py

@@ -130,7 +130,7 @@ def load_model(model_name):
 
             if args.gpu_memory:
                 params.append(f"max_memory={{0: '{args.gpu_memory or '99'}GiB', 'cpu': '{args.cpu_memory or '99'}GiB'}}")
-            elif (args.gpu_memory or args.cpu_memory) and not args.load_in_8bit:
+            elif not args.load_in_8bit:
                 total_mem = (torch.cuda.get_device_properties(0).total_memory/(1024*1024))
                 suggestion = round((total_mem-1000)/1000)*1000
                 if total_mem-suggestion < 800: