Update app.py

app.py

@@ -6,6 +6,7 @@ from transformers import (
     AutoModelForCausalLM,
     TextIteratorStreamer,
     pipeline,
+    AutoConfig,
 )
 from threading import Thread
 
@@ -21,13 +22,19 @@ from threading import Thread
 model_name_or_path = "TheBloke/phi-2-GPTQ"
 # To use a different branch, change revision
 # For example: revision="gptq-4bit-32g-actorder_True"
+
+config = AutoConfig.from_pretrained(model_name_or_path)
+config.quantization_config["disable_exllama"] = True
+
 model = AutoModelForCausalLM.from_pretrained(model_name_or_path,
                                              device_map="cpu",
                                              trust_remote_code=True,
-                                             revision="main")
+                                             revision="main",
+                                             config=config)
 
 tokenizer = AutoTokenizer.from_pretrained(model_name_or_path, use_fast=True)
 
+
 # Text generation pipeline
 phi2 = pipeline(
     "text-generation",