test app

app.py

@@ -33,23 +33,23 @@ from transformers import AutoModelForCausalLM, AutoTokenizer
 from transformers.generation import GenerationConfig
 
 
-DEFAULT_CKPT_PATH = 'WisdomShell/CodeShell-7B-Chat'
+DEFAULT_CKPT_PATH = 'WisdomShell/CodeShell-7B-Chat-int4'
 
 
 def _load_model_tokenizer(args):
     tokenizer = AutoTokenizer.from_pretrained(
-        'WisdomShell/CodeShell-7B-Chat', trust_remote_code=True, resume_download=True,
+        'WisdomShell/CodeShell-7B-Chat-int4', trust_remote_code=True, resume_download=True,
     )
 
     model = AutoModelForCausalLM.from_pretrained(
-        'WisdomShell/CodeShell-7B-Chat',
+        'WisdomShell/CodeShell-7B-Chat-int4',
         trust_remote_code=True,
         resume_download=True,
         torch_dtype=torch.bfloat16
     ).eval()
 
     config = GenerationConfig.from_pretrained(
-        'WisdomShell/CodeShell-7B-Chat', trust_remote_code=True, resume_download=True,
+        'WisdomShell/CodeShell-7B-Chat-int4', trust_remote_code=True, resume_download=True,
     )
 
     return model, tokenizer, config